Use of cone-beam computed tomography for investigation of factors affecting the failure of temporary anchorage devices

Abstract

Purpose: To assess the reasons behind the failure of temporary anchorage devices (TADs) installed in the buccal alveolar zone between the maxillary second premolars and the first molars.

Methods: Sixty patients (11 male, 49 female, mean age 21 years) were included. Logistic regression analysis was performed to determine the influence of the following factors on the presence or absence of TAD failure: sex, age, sagittal and vertical skeletal patterns, root contact, bone density, bone contact length, and presence of maxillary sinus perforation. Fisher’s exact test was performed to evaluate differences in failure rate between tooth types for teeth in contact with TADs (second premolar or first molar). Measurements were obtained using lateral cephalograms at the initial visit and data obtained by cone-beam computed tomography (CBCT) before and after TAD implantation.

Results: Logistic regression analysis showed that only root contact was associated with TAD failure; there was no association between the type of tooth in contact with the TAD and failure.

Conclusion: Root contact with a TAD significantly influenced TAD failure. Therefore, preliminary examination using three-dimensional (3D) evaluation with CBCT is essential to ensure that the TAD is not placed near the root of the tooth.

Introduction

Successful orthodontic treatment requires adequate anchorage control [1]. In the past, ancillary devices such as a Nance holding arch, transpalatal arch, and headgear were used in orthodontic treatment to strengthen the anchorage. However, it was difficult to fully control the anchorage as the mechanics were complicated and dependent on the patient’s level of cooperation [2]. Orthodontic miniscrew implants (temporary anchorage devices: TAD) are auxiliary devices that provide anchorage during tooth movement, preventing opposing tooth movements and thus allowing efficient and ideal tooth alignment [3]. In addition, as they can be implanted at various sites on the upper and lower jawbones with minimal surgical invasion, they have a wide range of indications and may be appropriate for difficult treatments, such as tooth intrusion and molar distalization [4,5]. They have several advantages in that they do not require supplementary orthodontic appliances such as headgear, and are not dependent on the patient’s level of cooperation, thus providing a high degree of reliability. Therefore, they have been widely and routinely used for orthodontic treatment in recent years.

However, the use of TADs is also associated with several problems, including damage to the periodontal ligament and roots, TAD breakage, and aphthous ulceration [6]. The most common problem is TAD failure, with an average TAD success rate of 87.21% (maxillary 89.87%, mandibular 79.24%) [7]. The success rate of dental implants is higher (>90%) [8]. Previous studies have reported that various factors such as age, clinical proficiency, cortical bone thickness, root contact, and bone density contribute to TAD failure [9,10,11].

As TADs are generally placed between adjacent teeth, they are proximal to the root. This may contribute to TAD failure. However, several reports have suggested that proximity to the root is not a determinant of TAD failure, and that TAD with periodontal ligament invasion without root contact can be successful if there is initial stability [11,12]. Ikenaka et al. have reported that the risk of TAD failure is 13 times higher when root contact is present, and that the failure rate becomes higher when the contact area between the root and the TAD increases. Previous studies have only evaluated the presence or absence of root proximity or root contact in relation to TAD failure; none have examined the relationship between the type of tooth contacted and TAD failure. The second molars are subjected to excessive posterior occlusal forces, and the percentage of occlusal forces is reported to be significantly higher posteriorly than anteriorly, thereby suggesting that the occlusal force applied by each tooth type differs [13]. It is clinically important to determine whether the type of tooth contacted is associated with TAD failure, and the present study was conducted to examine this issue.

TADs are made from a titanium alloy, and their primary aim is mechanical retention rather than osteointegration. The quality and quantity of bone at the implantation site, which vary according to sex, age, and vertical or sagittal skeletal patterns, are considered to be the main factors affecting TAD stability [14]. Vertical skeletal patterns are generally classified according to Frankfort-mandibular plane angle (FMA) into three types: hyperdivergent with a long vertical distance (FMA > 29°), hypodivergent with a short vertical distance (FMA < 21°), and normodivergent with a standard vertical distance (21° < FMA < 29°) [15]. Hyperdivergent and hypodivergent individuals are known to have thin and thick cortical bone, respectively [16]. The sagittal skeletal pattern is classified into three categories according to the angle between the A point, the nasion, and the B point (ANB): standard, maxillary prognathism, and mandibular prognathism, which are classified as skeletal class I (0° ≤ ANB < 6°), class II (ANB ≥ 6°), and class III (ANB < 0°) [17]. In general, individuals with maxillary prognathism have thicker cortical bone and greater cortical bone density [17,18]. Cortical bone is thicker in adults than in adolescents, and thicker and denser in males than in females [19,20]. Therefore, when placing a TAD, it is important to consider that bone density and cortical bone thickness vary according to sex, age, and skeletal pattern.

Cone-beam computed tomography (CBCT) is widely used in dentistry to acquire 3D images that cannot be obtained using conventional two-dimensional radiography techniques such as panoramic and dental radiography [21]. CBCT requires a lower radiation dose than multidetector computed tomography (MDCT). Developments in recent years mean that CBCT can be performed with even lower radiation doses and yields a wider useful radiation image. Therefore, it is used routinely in orthodontics to assess the morphologies of the jawbone and alveolar bone [22]. In addition, CBCT is being used increasingly to confirm the 3D positioning of TADs in relation to the jawbone and tooth root [23]. 3D evaluation of the implantation site using CBCT provides useful clinical information that reduces the risk of root contact for safe TAD placement.

There have been numerous attempts to evaluate the factors affecting the success rate of TADs, although most have involved animal studies and the focus was on implant stability immediately after placement rather than the clinical success rate [24,25]. In the context of anchoring screws, various factors have been reported to be associated with TAD failure, but the impact of each individual factor has not been investigated in detail. The aim of the present study was to examine the effect of multiple factors on TAD failure at the same time, allowing assessment of the contribution of each individual factor while considering other variables including age, sex, sagittal skeletal pattern, vertical skeletal pattern, root contact, bone contact length, and bone density in patients undergoing TAD placement between the maxillary second premolar and the first molar.

Materials and Methods

Prior to this study, a power analysis was performed using power analysis software (G*Power 3.19, Universität Düsseldorf, Düsseldorf, Germany) to estimate the sample size. In this investigation, the sample size was determined by setting the significance threshold at 0.05 and aiming for a power of 95%. According to the power analysis results, 80 TADs were needed in total. This study was approved by the Kanagawa Dental College Ethics Committee (Approval No. 889). Consent for participation was gathered through an opt-out approach. Any patients who chose not to participate were accordingly not included in the study.

The study sample comprised 60 patients (11 males, 49 females, mean age 21 years) who visited the orthodontic outpatient clinic of Kanagawa Dental College Hospital between December 2017 and September 2023 and underwent TAD placement (Vector TAS, Ormco Corporation, Orange, CA, USA). Each TAD was 8.0 mm in length and had a 1.4-mm outer diameter. A dental professional, with over three years of expertise in orthodontics, positioned it within the buccal alveolar area, between the maxillary first molar and the second premolar. CBCT images were taken before and after implant placement; such images were necessary for TAD placement and were not taken for research purposes.

CBCT (KaVo OP 3D Vision, KaVo Dental, Biberach, Germany) was used for imaging before TAD implantation with the following settings: tube voltage 120.0 kV, tube current 5.0 mA, slice width 0.3 mm, field of view (FOV) 170 × 230 mm, and imaging time 17.8 s. For imaging after TAD implantation, CBCT (3D Accuitomo, J. Morita, Kyoto, Japan) was used with the following settings: tube voltage 90.0 kV, tube current 5.0 mA, slice width 0.48 mm, FOV 60 × 60 mm, and imaging time 17.5 s. Data collected during the study were saved in the form of a Digital Imaging and Communications in Medicine (DICOM) file. The TAD implantation site was measured using CBCT image analysis software, InVivoDental6 (Anatomage, Santa Clara, CA, USA). All measurements were conducted by a single operator to ensure consistency and accuracy.

In total, 101 TADs were implanted. Successfully implanted TADs were defined as those that could be used as anchorage for at least 6 months after implantation; failed TADs were those that lost anchorage within the 6-month period after implantation [26]. Sagittal and vertical skeletal patterns were evaluated using the ANB and FMA, respectively, measured on the lateral cephalogram obtained at the initial visit (Fig. 1).

The TAD implantation site was reproduced by superimposing the CBCT data for the jawbone at the initial visit and immediately after TAD implantation using the superimposition function of InVivoDental6 (Fig. 2). The CBCT data at the initial examination were then adjusted in the axial and coronal sections to make them parallel to the long axis of the TAD; the predicted Hounsfield Unit value (predicted HU) was then measured (Fig. 3).

The predicted HU values were measured using the InVivoDental6 HU measurement tool, which displayed the maximum, minimum, and mean predicted HU values for the selected area. Maximum predicted HU values were selected on the basis of previous studies (Fig. 4) [Rossi et al., Int Orthod 15: 610-624, 2017.]. The predicted HU values were measured in the axial section over a range of 1.0 mm² for the cortical bone and 1 mm × length of implantation for the trabecular bone [20]. However, these values are difficult to measure accurately because the gray levels obtained by CBCT are not absolute values such as CT values, but rather relative values [27]. It has been reported that there is a correlation between HU values measured by CBCT and CT, and that HU units can be derived from gray levels with correction [28]. In this study, the correlation between the predicted HU values obtained by CBCT and MDCT was verified prior to measurement of bone mineral density [29]. The Spearman’s rank correlation coefficient between the predicted HU values obtained with CBCT and MDCT was r = 0.63, indicating a strong positive correlation between the two variables (P < 0.05). Therefore, the bone quality was considered evaluable.

The length of contact between the TAD and cortical and cancellous bone was determined from the CBCT data after implantation. Before measurement, the TAD was adjusted in the axial and coronal sections parallel to its long axis. The length of contact between cortical and cancellous bone and the TAD, the occurrence of contact with the root, specifically whether it was with the second premolar or the first molar, as well as the presence of maxillary sinus perforation, were measured. (Fig. 5).

The patients were then evaluated for sex, age, sagittal and vertical skeletal patterns, presence of maxillary sinus perforation, root contact and tooth type contacted, and cortical bone density. The following factors were evaluated for their influence on failure: cortical bone density, cancellous bone density, cortical bone contact length, and cancellous bone contact length.

To ensure the accuracy of measurements, a single evaluator randomly chose 20 subjects for reassessment 2 weeks after their first evaluation. Re-evaluation did not reveal any significant differences in the intraclass correlation coefficient (ICC) (P > 0.05), and also showed superior reliability of repeated 3D measurements performed on CBCT images with a 95% confidence interval. Park et al. have reported that the success rate is higher on the left side than on the right [30]. Because left-right differences may affect the failure rate, Student’s t-test and Fisher’s exact test were used to evaluate differences in the variables. The Mann-Whitney U test was then used to examine any differences in sex, age, sagittal and vertical skeletal patterns, root contact, bone density, bone contact length, and the presence of maxillary sinus perforation, and to determine the presence of any differences between the failure and non-failure groups in terms of each factor. The objective variable for this study was binary data indicating whether failure had occurred or not; logistic regression analysis was used to investigate the extent to which each factor influenced the presence or absence of failure. The variance inflation factors (VIFs) of all variables for the multicollinearity problem were confirmed to be less than 10. Fisher’s exact test was also performed to evaluate whether the failure rate differed significantly between the tooth types for teeth in contact with the TAD (maxillary second premolars or maxillary first molars). Statistical analyses were performed using statistical analysis software (IBM SPSS version 28.0.1, IBM Corp., Armonk, NY, USA). The significance level was set at α = 0.05 (two-tailed).

Fig. 1 a: vertical skeletal pattern (FMA), b: sagittal skeletal pattern (ANB).

Fig. 2 Superimposition of CBCT data (blue) for the jawbone at the initial visit (white) and after TAD placement

Fig. 3 a: Adjustment of the TAD angle to the long axis in axial section; b: Adjustment of the TAD angle to the long axis in coronal section.

Fig. 4 a: Measurement of trabecular bone density; b: Measurement of cortical bone density.

Fig. 5 a: Measurement of the length of contact between cortical and trabecular bone and TAD; b: Presence of root contact; c: Presence of maxillary sinus perforation

Results

Table 1 shows the characteristics of the participants (sex: 11 males, 49 females, total 60; age: 21 ± 7.0 years; skeletal pattern: sagittal 34 Class I, 15 Class II, 11 Class III; vertical 28 hyperdivergent, 27 normodivergent, and 5 hypodivergent).

The overall failure rate for TADs (51 on the right side and 50 on the left) placed in the buccal alveolar area between the maxillary second premolar and first molar was 15.8% (16 of 101 TADs). Student’s t-test and Fisher’s exact test were used to evaluate differences in the variables for each of the following factors: failure, sex, root contact, maxillary sinus perforation, sagittal and vertical skeletal patterns, and age. Fisher’s exact and Student’s t-tests were used to evaluate differences in the variables. Comparison of these factors between the left and right sides revealed no significant difference, and no left-right bias was found (Tables 2 and 3). Therefore, data from patients with TADs implanted on only one side were combined and analyzed. However, to solidify the findings, factors from both the right and left sides were incorporated into the analysis as variables.

The Mann-Whitney U test was used to investigate differences in any factors between the dropout and non-dropout groups. Among the factors age, sagittal and vertical skeletal patterns, cortical bone and cancellous bone densities, and cortical and cancellous bone contact lengths, only vertical skeletal pattern values showed a significant difference between the failure and non-failure groups (P = 0.024) (Table 4).

Logistic regression analysis was performed to investigate the influence of sex, age, sagittal skeletal pattern (ANB), vertical skeletal pattern (FMA), root contact, cortical bone and cancellous bone densities. Cortical and cancellous bone contact lengths, maxillary sinus perforation, and implantation site were examined to determine their influence on the presence or absence of failure. Among them, root contact alone showed a significant correlation (P = 0.001) with TAD failure risk, with an odds ratio of 28.25-fold (Table 5). Among patients in whom TADs did not fail, root contact was absent in 85.0%, whereas root contact was present in 75.0% of cases in which TAD failed. A statistically significant disparity was observed in the distribution (P < 0.001) (Table 6).

Fisher’s exact test was used to evaluate whether the failure rate differed significantly between tooth types for teeth in contact with the TAD (maxillary second premolars or maxillary first molars). Among the root-contacting TADs, two cases were failures, while six cases were successful (i.e. two out of eight were failures) when contacting the maxillary second bicuspids. The number of failures in cases of contact with the maxillary first molars was 10, and the number of successes was seven (i.e. 10 failures out of 17 cases). In total, 83.3% of the root-contacting and failed TADs were in contact with maxillary first molars. However, the results of Fisher’s exact test indicated no statistical significance (P = 0.202) (Table 6).

Table 1 Characteristics of the study participants

Characteristics	Number ( % )
Sex (male/female)	11/49
Age (years ± SD)	21.0 ± 7.0
Sagittal skeletal pattern
Class Ⅰ	34 (56.6)
Class Ⅱ	15 (25.0)
Class Ⅲ	11 (18.3)
Vertical skeletal pattern
hyperdivergent	28 (46.6)
normodivergent	27 (45.0)
hypodivergent	5 (8.3)
Total	60 (100)

Table 2 Right-left differences (TAD failure, sex, root contact, maxillary sinus perforation)

			Pooled	Failure		Sex		Root contact		Maxillary sinus perforation
				yes	no	male	female	with	without	with	without
Implantation side	left	n	50	5	45	9	41	13	37	4	46
		rate (%)		10.0	90.0	18.0	82.0	26.0	74.0	8.0	92.0
	right	n	51	11	40	8	43	12	39	2	49
		rate (%)		21.6	78.4	15.7	84.3	23.5	76.5	3.9	96.1
Total		n	101	16	85	17	84	25	76	6	95
		rate (%)		15.8	84.2	16.8	83.2	24.8	75.2	5.9	94.1
P value				0.17		0.80		0.82		0.44

Table 3 Right-left differences (age, and sagittal and vertical skeletal patterns)

		Value	Mean	SD	P
Age (years)	left	50	19.9	5.0	0.26
	right	51	21.3	7.4
	total	101	20.6	6.3
ANB (°)	left	50	3.6	3.3	0.96
	right	51	3.6	3.5
	total	101	4.4	7.2
FMA (°)	left	50	30.4	7.0	0.91
	right	51	30.2	6.7
	total	101	29.5	7.0

Table 4 Differences between failure and success groups (age, skeletal pattern, bone density, and bone contact length)

	Failure (no)	Failure (yes)	P
	mean ± SD (min/max)	(min/max)
Age (years)	20.3 ± 5.2 (12/40)	22.3 ± 10.6 (14/58)	0.94
ANB (°)	3.5 ± 3.4 (−4.9/9.2)	3.9 ± 3.4 (−2.1/10.4)	0.94
FMA (°)	29.7 ± 6.8 (19.5/49.0)	33.4 ± 6.1 (23.9/41.0)	0.02
Cortical bone density (predicted HU)	850.2 ± 178.7 (492.0/1260.0)	856.0 ± 137.2 (533.0/1059.0)	0.76
Cancellous bone density (predicted HU)	529.6 ± 149.8 (231.0/939.0)	529.6 ± 171.1 (302.0/826.0)	0.91
Cortical bone contact length (mm)	1.3 ± 0.2 (1.0/2.0)	1.3 ± 0.3 (1.0/1.8)	0.78
Cancellous bone contact length (mm)	5.3 ± 1.0 (1.6/7.9)	4.9 ± 1.1 (2.7/6.8)	0.13

Table 5 Association of TAD failure with various factors (side of placement, sex, age, skeletal pattern, bone density, bone contact length, root contact, maxillary sinus perforation)

	B	SE	Wald	df	Sig	Exp (B)	Exp (B)(95%Cl)
							min	max
Implantation side	1.28	0.81	2.48	1	0.12	3.58	0.73	17.52
Sex (male/female)	−0.43	1.14	0.14	1	0.71	0.65	0.07	6.05
Age (years)	0.08	0.05	3.01	1	0.08	1.08	0.99	1.19
ANB (°)	−0.05	0.13	0.17	1	0.68	0.95	0.74	1.22
FMA (°)	0.07	0.06	1.37	1	0.24	1.08	0.95	1.21
Cortical bone density (predicted HU)	0.00	0.00	0.46	1	0.50	1.00	1.00	1.01
Cancellous bone density (predicted HU)	0.00	0.00	1.40	1	0.24	1.00	0.99	1.00
Cortical bone contact length (mm)	0.29	1.71	0.03	1	0.87	1.34	0.05	37.86
Cancellous bone contact length (mm)	−0.18	0.32	0.33	1	0.56	0.83	0.45	1.55
Root contact (yes/no)	3.34	0.86	15.26	1	0.00	28.25	5.29	151.04
Maxillary sinus perforation	−17.84	15839.05	0.00	1	1.00	0.00	0.00
Constant	−6.67	4.57	2.13	1	0.14	0.00

B: regression coefficient; SE: standard error of B; Wald: wald chi-squared value used to test the significance of B; df: degrees of freedom; Sig: P-value indicating statistical significance; Exp(B): exponentiation of the B coefficient, indicating the odds ratio; 95% CI for Exp(B): 95% confidence interval for the odds ratio; FMA: frankfort-mandibular plane angle; HU: hounsfield unit; ANB: angle between the A point, nasion, and B point

 

Table 6 Differences between failure and success groups (root contact, type of tooth contacted, sex, maxillary sinus perforation)

		n	Failure		P
			no	yes
Root contact	no	76	72	4	0.001
	yes	25	13	12
	total	101	85	16
Contacting tooth	second premolar	8	6	2	0.202
	first molar	17	7	10
	total	25	13	12
Sex	male	17	15	2	1.000
	female	84	70	14
	total	101	85	16
Maxillary sinus perforation	no	95	79	16	0.586
	yes	6	6	0
	total	101	85	16

Discussion

In this study, root contact was most strongly associated with TAD failure, being consistent with the results of most previous studies. Primary stability is important for TAD stability; it is the initial holding force achieved by mechanical retention between the bone and the TAD. Proximity or contact of the TAD with the tooth root causes occlusal forces to be transmitted through the tooth to the TAD, inhibiting the process of bone reconstruction around the latter [31]. It has been suggested that this leads to inadequate bone remodeling, resulting in a smaller surrounding bone mass than for a properly placed TAD [32]. Any slight movement of the TAD after implantation may lead to invasion of fibrous tissue, thereby promoting failure. Therefore, TAD and root contact are considered to have a significant influence on stability.

To prevent root contact with the TAD, it is important to consider the site where TAD placement has sufficient space, and to select an appropriately shaped TAD. A sufficient amount of interdental bone is present proximal to the maxillary first molar to allow implantation of a TAD [33]. It has been reported that at least 1.4 mm of bone is required on both sides of the TAD to prevent proximity to the root and increase stability [34]. Some reports suggest that root contact is likely to be avoided by using a TAD with an external dimension of less than 1.5 mm, as there is at least 1.5 mm between the maxillary and maxillary roots [35]. However, unlike TAD placement in the palate, the risk of root contact cannot be eliminated when a TAD is placed between the roots. Therefore, the risk of root contact can be reduced by confirming the presence of sufficient space at the implantation site through 3D evaluation using CBCT prior to TAD implantation. The length of the TAD is determined based on the condition of the mucosa at the site of implantation. Longer TADs are recommended in areas with thick mucosa. In the upper and lower buccal alveolar regions, the usual 6-8 mm can be used to ensure safety [34]. Therefore, the TAD selected in this study, with an outer diameter of 1.4 mm and a length of 8.0 mm, is considered to have a minimal risk of root contact and can easily achieve stability after implantation.

During orthodontic treatment, tooth mobility increases due to bone resorption caused by expansion of the periodontal ligament space and bone remodeling [36]. Occlusal forces cause the teeth to tilt in a proximal direction. Therefore, when the TAD contacts the root of a tooth that is located more centrally than the implantation site, occlusal forces are likely to be transmitted to the TAD, which may interfere with primary stability. In this study, we investigated the effect of the contact tooth type on failure. There was a tendency for TADs in contact with the maxillary first molars to fail compared to the maxillary second bicuspids, but the difference was not statistically significant. In contrast, the number of root-contacting TADs used in the analysis was 26 out of 101. The results of the post-hoc power analysis indicated that the power was approximately 0.37, suggesting that the present sample size may not have had sufficient power to detect differences between tooth types. It is also difficult to predict the forces applied to different tooth types, owing to the differences in occlusal forces between malocclusion types and changes in occlusal contact with the progression of bicuspid extraction and orthodontic treatment. Therefore, in future studies, it would be important to consider appropriate sample size, measurement of occlusal force and occlusal contact for each tooth type, and evaluation of tooth mobility before and after implantation; further studies should take these factors into consideration.

The primary stability of a TAD is achieved by mechanical contact between the TAD surface and the bone, and therefore dependent on the quality and quantity of bone at the implantation site [20]. The use of CT scans has become common to objectively determine bone density at implant placement sites and to assess bone quality using HU; however, HU bone density values on CBCT images are reportedly unreliable because they are higher than those obtained on MDCT images. One study has suggested that there is a correlation between the predicted HU value measured by CBCT and the HU value measured by CT, and that the gray level obtained from dental CBCT can be corrected using a linear attenuation factor to approximate the HU value obtained by MDCT [31]. Therefore, predicted HU values measured by CBCT and those measured by MDCT cannot be treated as synonymous; however, it is possible to compare the magnitude of predicted HU values for each sample. No significant variance in bone density was observed between TADs that succeeded and those that failed. Hence, the present findings suggest that bone density might not be the principal element influencing the failure of TADs.

Mechanical contact is an important factor influencing the primary stability of a TAD, and the length of contact with the cortical and trabecular bone has been reported to be closely related to mechanical contact. Factors that influence cortical bone thickness include age, sex, and vertical and sagittal skeletal patterns. Logistic regression analysis did not reveal any association between the length of contact with the cortical or trabecular bone and failure. The average contact length of the cortical bone was 1.29 mm and that of the trabecular bone was 5.31 mm. Some reports have indicated that implantation should be performed at a site with a cortical bone thickness of 1 mm or greater to increase the success rate [37]. The smallest cortical bone contact length observed in this study was 1.03 mm, suggesting that a sufficient amount of cortical bone was available. Thus, the presence of more than 1 mm of cortical bone may obscure the association between the bone contact length and TAD failure. However, a statistically significant difference was observed solely in the vertical skeletal patterns of TADs that were successful compared with those that failed. Vertical skeletal patterns have been reported to affect cortical bone thickness, suggesting that latter may influence TAD stability.

The discrepancy between the results of comparative analysis (Mann-Whitney U test) and logistic regression analysis may have been due to the latter’s ability to account for confounding factors. Logistic regression analysis controls for multiple variables simultaneously, thereby isolating the effect of each variable on TAD failure. In this study, while the Mann-Whitney U test indicated a significant association between vertical skeletal pattern and TAD success, the logistic regression analysis did not show such an association. This suggests that when confounding factors are controlled for, the vertical skeletal pattern alone does not significantly affect TAD failure.

As the number of failed TADs was 13 on the left side and 12 on the right, and no effect of the left-right difference on failure was observed, the data for patients who had TADs implanted on only one side were also included in the analysis. In the past, it has been reported that inflammation around the TAD is associated with the TAD failure rate [6]. The influence of the dominant hand of the implanting operator on the left-right difference has also been considered: Kravitz et al. have reported that operators tend to inadvertently pull the hand driver towards the body during TAD placement to change the angle of the TAD, thus increasing the potential for contact with the root [6]. Therefore, it was considered essential to operate the hand driver at a slight distance from the body during each rotation and to employ a finger rest for the implantation angle or a surgical guide while inserting the TAD. This study did not include information on the dominant hand or soft tissue of the surgeon. Future evaluations of the influence of these factors on TAD failure may provide clinically useful information.

At the location of the first molar, the sinus floor reaches its greatest depth, and maxillary sinus perforation can occur when a TAD is placed between the maxillary second premolar and first molar. Small (<2 mm) perforations of the maxillary sinus heal spontaneously without complications and do not require immediate removal [6]. It was considered that reduced bone contact length due to maxillary sinus perforation might be a factor related to TAD failure; however, no association between maxillary sinus perforation and TAD failure was found. It has been reported that there is no difference in stability between implants that perforate the nasal cavity and maxillary sinus and those that do not [38]. This is consistent with the present results. Therefore, although it seems that maxillary sinus perforation was not associated with failure, possible complications should be considered.

This study had several limitations. First, the overall failure rate was 15.0% (15 of 100 TADs), suggesting that the proportion of TADs that failed was small, and it may have been difficult to detect factors associated with failure. Further studies of the factors related to TAD failure employing a sufficient sample size are therefore warranted. Information on the cleaning status of periodontal tissues and the presence of inflammation was not verified. The measurement of bone density using CBCT may also have been inaccurate. Considering these limitations, future studies should include a more detailed analysis of the various factors that may affect the stability and success rate of TADs, such as the periodontal tissue conditions and the use of techniques that allow for more accurate measurement of bone density.

The present study showed that root contact of the TAD significantly affected its likelihood of failure. Therefore, it is essential to ensure that TADs are not placed in proximity to the root using CBCT or CT 3D evaluation in preliminary diagnosis.

Abbreviations

CBCT: cone-beam computed tomography; MDCT: multidetector computed tomography; DICOM: digital imaging and communications in medicine; FMA: Frankfort-mandibular plane angle; VIF: variance inflation factor

Ethical Statements

This study was approved by the Kanagawa Dental College Ethics Committee (Approval No. 889). Informed consent was obtained in the form of opt-out. Patients who refused to participate were excluded.

Conflicts of Interest

The authors have no conflicts of interest to declare.

Funding

Not applicable

Author Contributions

TY: conceptualization, methodology, project administration, resources, writing-review and editing; AY: data curation, formal analysis, investigation, software, validation, visualization, writing-original draft; SK: formal analysis, methodology, software, supervision, writing-review and editing; RI: validation, writing-review and editing. All authors read and approved the final version of the manuscript.

ORCID iD

AY: [e-mail], NA

SK*: [e-mail], https://orcid.org/0000-0001-6552-6925

RI: [e-mail], NA

TY: [e-mail], https://orcid.org/0000-0001-9806-7163

Acknowledgments

The authors thank Dr. Takumi Imai at the Clinical and Translational Research Center, Kobe University Hospital, for his advice on statistical analysis and Dr. Masahiro Izumi at the Department of Diagnostic Imaging, Kanagawa Dental University Hospital, for his advice regarding the study.

Data Availability Statements

The data supporting the findings of this study are fully available within the article.

References

• 1.  Khlef HN, Hajeer MY, Ajaj MA, Heshmeh O (2018) Evaluation of treatment outcomes of en masse retraction with temporary skeletal anchorage devices in comparison with two-step retraction with conventional anchorage in patients with dentoalveolar protrusion: a systematic review and meta-analysis. Contemp Clin Dent 9, 513-523. doi: 10.4103/ccd.ccd_661_18
• 2.  Feldmann I, Bondemark L (2008) Anchorage capacity of osseointegrated and conventional anchorage systems: a randomized controlled trial. Am J Orthod Dentofacial Orthop 133, 339.e19-e28. doi: 10.1016/j.ajodo.2007.08.014
• 3.  Baxi S, Bhatia V, Tripathi A, Prasad Dubey M, Kumar P, Mapare S (2023) Temporary anchorage devices. Cureus 15, e44514. doi: 10.7759/cureus.44514
• 4.  Yáñez-Vico RM, Iglesias-Linares A, Cadenas de Llano-Pérula M, Solano-Reina A, Solano-Reina E (2014) Management of occlusal canting with miniscrews. Angle Orthod 84, 737-747. doi: 10.2319/051313-369.1
• 5.  Ramsundar K, Jain RK, Balakrishnan N (2023) Dentoalveolar and airway changes following en masse distal movement of the maxillary dentition with infrazygomatic crest anchorage: a prospective study. J Int Soc Prev Community Dent 13, 62-67. doi: 10.4103/jispcd.JISPCD_193_22
• 6.  Kravitz ND, Kusnoto B (2007) Risks and complications of orthodontic miniscrews. Am J Orthod Dentofacial Orthop 131, S43-S51. doi: 10.1016/j.ajodo.2006.04.027
• 7.  Lee DW, Park JH, Bay RC, Choi SK, Chae JM (2021) Cortical bone thickness and bone density effects on miniscrew success rates: a systematic review and meta-analysis. Orthod Craniofac Res 24, 92-102. doi: 10.1111/ocr.12453
• 8.  Gokcen-Rohlig B, Yaltirik M, Ozer S, Tuncer ED, Evlioglu G (2009) Survival and success of ITI implants and prostheses: retrospective study of cases with 5-year follow-up. Eur J Dent 3, 42-49. doi: 10.1055/s-0029-1197085
• 9.  Chen YJ, Chang HH, Huang CY, Hung HC, Lai EH, Yao CC (2007) A retrospective analysis of the failure rate of three different orthodontic skeletal anchorage systems. Clin Oral Implants Res 18, 768-775. doi: 10.1111/j.1600-0501.2007.01405.x
• 10.  Lee MY, Park JH, Kim SC, Kang KH, Cho JH, Cho JW et al. (2016) Bone density effects on the success rate of orthodontic microimplants evaluated with cone-beam computed tomography. Am J Orthod Dentofacial Orthop 149, 217-224. doi: 10.1016/j.ajodo.2015.07.037
• 11.  Ikenaka R, Koizumi S, Otsuka T, Yamaguchi T (2022) Effects of root contact length on the failure rate of anchor screw. J Oral Sci 64, 232-235. doi: 10.2334/josnusd.21-0536
• 12.  Janson G, Gigliotti MP, Estelita S, Chiqueto K (2013) Influence of miniscrew dental root proximity on its degree of late stability. Int J Oral Maxillofac Surg 42, 527-534. doi: 10.1016/j.ijom.2012.09.010
• 13.  Qadeer S, Yang L, Sarinnaphakorn L, Kerstein RB (2016) Comparison of closure occlusal force parameters in post-orthodontic and non-orthodontic subjects using T-Scan® III DMD occlusal analysis. Cranio 34, 395-401. doi: 10.1080/08869634.2015.1122277
• 14.  Chugh T, Jain AK, Jaiswal RK, Mehrotra P, Mehrotra R. (2013) Bone density and its importance in orthodontics. J Oral Biol Craniofac Res 3, 92-97. doi: 10.1016/j.jobcr.2013.01.001
• 15.  Fawaz P, Amm E, Khoury E. (2021) The curve of Spee and its levelling in different vertical skeletal patterns: a 3D set-up study. Int Orthod 19, 659-668. doi: 10.1016/j.ortho.2021.08.003
• 16.  Ozdemir F, Tozlu M, Germec-Cakan D (2013) Cortical bone thickness of the alveolar process measured with cone-beam computed tomography in patients with different facial types. Am J Orthod Dentofacial Orthop 143, 190-196. doi: 10.1016/j.ajodo.2012.09.013
• 17.  Deguchi T Sr, Katashiba S, Inami T, Foong KW, Huak CY (2010) Morphologic quantification of the maxilla and the mandible with cone-beam computed tomography. Am J Orthod Dentofacial Orthop 137, 218-222. doi: 10.1016/j.ajodo.2008.02.029
• 18.  Khumsarn N, Patanaporn V, Janhom A, Jotikasthira D (2016) Comparison of interradicular distances and cortical bone thickness in Thai patients with Class I and Class II skeletal patterns using cone-beam computed tomography. Imaging Sci Dent 46, 117-125. doi: 10.5624/isd.2016.46.2.117
• 19.  Farnsworth D, Rossouw PE, Ceen RF, Buschang PH (2011) Cortical bone thickness at common miniscrew implant placement sites. Am J Orthod Dentofacial Orthop 139, 495-503. doi: 10.1016/j.ajodo.2009.03.057
• 20.  Cassetta M, Sofan AA, Altieri F, Barbato E (2013) Evaluation of alveolar cortical bone thickness and density for orthodontic mini-implant placement. J Clin Exp Dent 5, e245-e252. doi: 10.4317/jced.51228
• 21.  Friedlander-Barenboim S, Hamed W, Zini A, Yarom N, Abramovitz I, Chweidan H et al. (2021) Patterns of cone-beam computed tomography (CBCT) utilization by various dental specialties: a 4-year retrospective analysis from a dental and maxillofacial specialty center. Healthcare (Basel) 9, 1042. doi: 10.3390/healthcare9081042
• 22.  Kapila SD, Nervina JM (2015) CBCT in orthodontics: assessment of treatment outcomes and indications for its use. Dentomaxillofac Radiol 44, 20140282. doi: 10.1259/dmfr.20140282
• 23.  Chawshli OF, Hasan HS, Yalda FA, Al-Talabani SZ (2024) The success rate of mini-screws for incisors intrusion and patient age, gender, and insertion angle in the maxilla using CBCT and implant-guided surgery. Orthod Craniofac Res 27, 118-125. doi: 10.1111/ocr.12700
• 24.  Marquezan M, Souza MM, Araújo MT, Nojima LI, Nojima Mda C (2011) Is miniscrew primary stability influenced by bone density? Braz Oral Res 25, 427-432. doi: 10.1590/s1806-83242011000500009
• 25.  Yu WP, Tsai MT, Yu JH, Huang HL, Hsu JT (2022) Bone quality affects stability of orthodontic miniscrews. Sci Rep 12, 2849. doi: 10.1038/s41598-022-06851-y
• 26.  Suzuki M, Deguchi T, Watanabe H, Seiryu M, Iikubo M, Sasano T et al. (2013) Evaluation of optimal length and insertion torque for miniscrews. Am J Orthod Dentofacial Orthop 144, 251-259. doi: 10.1016/j.ajodo.2013.03.021
• 27.  Reeves TE, Mah P, McDavid WD (2012) Deriving Hounsfield units using grey levels in cone beam CT: a clinical application. Dentomaxillofac Radiol 41, 500-508. doi: 10.1259/dmfr/31640433
• 28.  Mah P, Reeves TE, McDavid WD (2010) Deriving Hounsfield units using grey levels in cone beam computed tomography. Dentomaxillofac Radiol 39, 323-335. doi: 10.1259/dmfr/19603304
• 29.  Cassetta M, Stefanelli LV, Pacifici A, Pacifici L, Barbato E (2014) How accurate is CBCT in measuring bone density? A comparative CBCT-CT in vitro study. Clin Implant Dent Relat Res 16, 471-478. doi:10.1111/cid.12027
• 30.  Park HS, Jeong SH, Kwon OW (2006) Factors affecting the clinical success of screw implants used as orthodontic anchorage. Am J Orthod Dentofacial Orthop 130, 18-25. doi: 10.1016/j.ajodo.2004.11.032
• 31.  Kuroda S, Yamada K, Deguchi T, Hashimoto T, Kyung HM, Takano-Yamamoto T (2007) Root proximity is a major factor for screw failure in orthodontic anchorage. Am J Orthod Dentofacial Orthop 131, S68-S73. doi: 10.1016/j.ajodo.2006.06.017
• 32.  Watanabe H, Deguchi T, Hasegawa M, Ito M, Kim S, Takano-Yamamoto T (2013) Orthodontic miniscrew failure rate and root proximity, insertion angle, bone contact length, and bone density. Orthod Craniofac Res 16, 44-55. doi: 10.1111/ocr.12003
• 33.  Schnelle MA, Beck FM, Jaynes RM, Huja SS (2004) A radiographic evaluation of the availability of bone for placement of miniscrews. Angle Orthod 74, 832-837. doi: 10.1043/0003-3219(2004)074<0832:AREOTA>2.0.CO;2
• 34.  Deguchi T, Nasu M, Murakami K, Yabuuchi T, Kamioka H, Takano-Yamamoto T (2006) Quantitative evaluation of cortical bone thickness with computed tomographic scanning for orthodontic implants. Am J Orthod Dentofacial Orthop 129, 721.e7-e12. doi: 10.1016/j.ajodo.2006.02.026
• 35.  Motoyoshi M (2011) Clinical indices for orthodontic mini-implants. J Oral Sci 53, 407-412. doi: 10.2334/josnusd.53.407
• 36.  Tanaka E, Ueki K, Kikuzaki M, Yamada E, Takeuchi M, Dalla-Bona D et al. (2005) Longitudinal measurements of tooth mobility during orthodontic treatment using a periotest. Angle Orthod 75, 101-105. doi: 10.1043/0003-3219(2005)075<0101:LMOTMD>2.0.CO;2
• 37.  Motoyoshi M, Yoshida T, Ono A, Shimizu N (2007) Effect of cortical bone thickness and implant placement torque on stability of orthodontic mini-implants. Int J Oral Maxillofac Implants 22, 779-784.
• 38.  Ardekian L, Oved-Peleg E, Mactei EE, Peled M (2006) The clinical significance of sinus membrane perforation during augmentation of the maxillary sinus. J Oral Maxillofac Surg 64, 277-282. doi: 10.1016/j.joms.2005.10.031