(1) To report the thickness of the cortical bone in insertion sites commonly used for orthodontic mini-implants, (2) to assess the impact of a change in insertion angle on primary cortical bone-to-implant contact, and (3) to evaluate the risk of maxillary sinus perforation.
At autopsy, 27 human samples containing three to five adjacent teeth were excised and scanned using a table-top micro-computed tomography system. Bone thickness measurements were taken at 45° and 90° to the long the axis of the adjacent teeth, simulating a mini-implant insertion at the mid-root level.
In the maxilla, the overall mean cortical thickness at 90° was 0.7 mm buccally in the lateral region, 1.0 mm buccally in the anterior region, and 1.3 mm palatally. In the mandible, the mean cortical thickness was 0.7 mm buccally and 1.8 mm lingually in the anterior region; 1.9 mm buccally and 2.6 mm lingually in the lateral region. Changing the insertion angle from 90° to 45° increased the cortical bone-to-implant contact by an average of 47%. Perpendicular insertion at the mid-root level only rarely interfered with the sinus, whereas apically inclined insertion increased the risk of sinus perforation.
Buccally and palatally in the maxilla and buccally in the anterior mandible, the thickness of the alveolar cortical bone is often less than 1 mm. In contrast, the alveolar cortical bone is frequently thicker than 2 mm laterally in the mandible. Changing the insertion angle to 45° will generally enhance implant stability but increase the risk of perforation to the maxillary sinus.
Primary stability of skeletal anchorage is dependent on the quantity and quality of bone in the insertion site. It has been reported that the cortical bone should have a thickness of more than 1 mm in order to obtain good stability of orthodontic mini-implants.1–3 The most used and easily accessible insertion sites are the buccal aspect of the alveolar process in both the maxilla and mandible as well as the palatal side of the maxillary alveolar process in the premolar and molar region.4 The thickness of human cortical bone in these areas has been assessed by conventional computed tomography (CT)5–8 and cone-beam computed tomography (CBCT).9–13 More detailed information can be achieved using micro-computed tomography (micro-CT). Indeed, compared to CBCT and CT, micro-CT scanning offers an enhancement of voxel quality and a reduction in voxel size, providing higher precision imaging.14 A detailed description of the alveolar process might provide the clinician with fundamental knowledge, increasing the success of orthodontic treatment with mini-implant anchorage. To the knowledge of the authors, the entire human dentoalveolar bone complex has not been studied previously by means of micro-CT.
The general recommendation is to place mini-implants in attached gingiva,4,15,16 yet as apical as possible since the interradicular distance increases in the apical direction, reducing the risk of root damage.13,16 A solution is to insert the mini-implants at an angle directing the mini-implant apically, which will also lead to an increased primary bone-to-implant contact.5,17 On the other hand, this solution might increase the risk of sinus perforation. In particular, the risk of sinus perforation with interradicular mini-implants placed from the buccal aspects with an apical insertion direction so far has not been addressed systematically. The risk of perforation of the maxillary sinus during mini-implant insertion has merely been reported in relation to insertion at the apical level of the molars, including the infrazygomatic crest and distally to the second molar.9,10,17–20 Long mini-implants of 8 to 10 mm have been reported to be associated with a higher risk of sinus membrane perforation than the shorter mini-implants of 6 mm.21
The purpose of this study was to: (1) provide detailed information of the cortical thickness in the most commonly used insertion sites, (2) assess how a change in the insertion angle influences the primary cortical bone-to-implant contact, and (3) evaluate the risk of sinus perforation during mini-implant insertion.
MATERIALS AND METHODS
Permission from the National Committee on Health Research Ethics was granted to collect tissue samples containing teeth and surrounding alveolar bone at human autopsy of 27 adult donors (18 male, 9 female). The age of the donors ranged from 20 to 50 years, with a mean age of 34 years. None of the donors had report of disease or medication, which could influence the bone turnover. A dentoalveolar bone-block containing three to five adjacent teeth was excised from each individual. All teeth were in occlusion, and except from third molars, all teeth were represented in at least four samples. No differentiation between the right and left side was made. The samples were fixed in 70% alcohol before micro-CT scanning was performed with a table-top micro-CT system (μCT40, Scanco Medical, Bassersdorf, Switzerland) at an energy of 70 kVp and a current of 113 µA. A standard-medium resolution mode (1024 pixels, 500 projections) with an integration time of 200 ms was chosen, resulting in an isotropic voxel of 37 µm and an in-plane dimension of 1024 × 1024 pixels. To improve the validity and quality of the reconstructed images, an oversampling factor of 3 was adopted during the scanning (Figure 1). After scanning, the single-slice images were exported as stacks of TIFF files.
The TIFF file datasets were imported into Mimics (Materialise, Leuven, Belgium), where all of the measurements were performed. The datasets were aligned in the three planes of space along the tooth axis of the adjacent teeth. Before every set of measurements, the root length of each tooth, defined as the distance from the cementoenamel junction to the apex, was measured. One hundred fifty-one possible insertion sites were examined. The thickness of the cortical bone was measured buccally as well as palatally/lingually, corresponding to the approximal spaces between all teeth. To simulate an insertion in the attached gingiva close to the mucogingival junction, the measurements were performed at 45° and 90° in relation to the long axis of the teeth at the mid-root level (Figure 2). At the same level, the risk of perforating to the maxillary sinus (Figure 3a) was assessed by measuring the distance from the outer aspect of the palatal and buccal cortical bone plate to the maxillary sinus cavity at 45° and 90° (Figure 3b). It was noted whether the simulated insertion path caused a sinus perforation or not.
The three-dimensional (3D) rendering of the samples was performed after importing the datasets into VGStudio MAX (Volume Graphics GmbH, Heidelberg, Germany).
The data were grouped according to localization in either the upper or lower jaw. Each group was subdivided with respect to buccal and palatal/lingual aspect and according to the insertion angle. Mean values were calculated for each group and presented along with the minimum and maximum value. The error of the method for determination of cortical thickness was calculated by double measurements of 10 randomly selected sites from 10 randomly selected samples using the Dahlberg formula (s = √Σd2 / 2n, where d = difference between the first and second measurements).22 The error of the method for the distance to the sinus from the outer (ie, buccal and palatal) cortical bone was calculated on double measurements of 10 randomly selected sites in the four samples available.
Error of the Method
The method error for the measurement of the cortical bone thickness was 0.15 mm at 90° and 0.27 mm at 45°. The method error for the distance to the maxillary sinus from the outer cortical bone was 0.63 mm at 45°, while at 90° it was not calculated because only one interdental site presented with risk of sinus perforation at this insertion angulation.
Thickness of Cortical Bone at 90°
The mean thickness of the buccal cortical bone posterior to the lateral incisor in the maxilla and anterior to the first premolar in the mandible was less than 1 mm measured perpendicularly to the long axis of the teeth (Table 1A,B). Between the maxillary incisors the mean buccal cortical thickness was 1.1 mm. Palatally, the cortical bone was on average slightly thicker with an overall mean value of 1.3 mm. The cortical bone was found laterally in the mandible where the cortex was thicker both lingually and buccally than in any other region. The average cortical bone thickness laterally in the mandible ranged from 1.45 mm to 2.99 mm, at an angle of 90°. Lingually in the anterior mandible, the mean cortical bone was 1.81 mm thick. In both jaws, the cortical bone was generally thicker on the lingual/palatal side than on the corresponding buccal side. A large interindividual variation was noted.
Impact of a Change in Insertion Angle to 45°
Risk of Sinus Perforation
The maxillary sinus floor extended further into the alveolar process in relation to the molars than to the premolars, but rarely reached the mid-root level (Table 2). Therefore, when insertion was simulated perpendicular to the long axis of the teeth, only one interdental space involved a risk of sinus perforation. A change in the angle to 45°, on the other hand, did increase the risk of perforation considerably. Examining the interdental spaces from second premolar to second molar, all interdental spaces presented with risk of perforation at a 45° insertion angle. At 45°, the distance from the outer-buccal or outer-palatal cortical bone plate to the sinus was ranging from 2.1 to 8.8 mm. A thin layer of cortical bone lining the maxillary sinus was found in all four samples.
The aim of this study was to analyze the anatomy of the human alveolar bone by means of high-precision 3D micro-CT scanning of human autopsy material. In particular, the thickness of the cortical bone was assessed in the most common insertion sites of mini-implants, as the primary stability of mini-implant systems is first and foremost dependent on the bone quantity and quality of the insertion site.1,3 Because the samples derived from autopsy material, alveolar bone blocks could be harvested, and a high-resolution micro-CT scanning could be applied. This resulted in an isotropic voxel size of 0.037 mm with a minimized partial volume effect compared to CT and CBCT (in the most recent CBCT study10 on cortical thickness the voxel size was 0.28 mm). The voxel size has a critical influence on the measured thickness of the cortical bone: Tsutsumi et al.23 demonstrated in an in vitro CBCT study that the measured structure must have a thickness of at least 3 to 4 voxel size to maintain high accuracy of the measurements. The thickness of the alveolar cortical bone was found to be often less than 1 mm buccally in the entire maxilla and anteriorly in the mandible, corroborating earlier reports.5,7,10–12 Thus, the accuracy of the micro-CT measurements was high compared to earlier studies. On the other hand, the limited sample size of the present study was a limitation, especially considering the large interindividual variation.
Different references have been used for measuring the thickness of the cortical bone, eg, the palatal plane,12 the Frankfort horizontal plane,13 the occlusal plane,7 the mandibular plane,8 the perpendicular to the bone surface,10,11 of the local insertion site and the long axis of adjacent teeth.5,6 We opted for the latter method because it seems to be clinically applicable. However, the direction of the measurements has an influence on the measurement of the cortical thickness, as the surface is generally slightly angulated in relation to the long axis of the teeth. The use of different measurement and scanning methods might explain some of the variation of the measured cortical bone thickness found in the literature.5–8,10–13 Additionally, the intraindividual irregular thickness of the cortical bone seen in Figure 2 increases the variation of the measured thickness.
The marked range between minimum and maximum values of the cortical bone thickness found between same sites in different samples could be ascribed to a large interindividual variation also noted by Baumgaertel10 and Baumgaertel and Hans.11 Taking into account the mean values and the large variation, it can be deducted that the thickness of the cortical bone in the individual patient often will be less than 1 mm at the midroot level both buccally and palatally in the entire maxillary alveolar process, as well as buccally in the anterior mandibular alveolar process. Therefore, the stability of the mini-implants eventually has to rely on the trabecular bone structure in these areas1; consequently, a long mini-implant with an intraosseous length of 8 mm might be the right choice.1 Moreover, an apical directed insertion will increase the primary cortical bone-to-implant contact. On the other hand, the alveolar cortical bone buccally and lingually in the posterior mandible would frequently be thicker than 2 mm, regardless of the insertion angle. In this region, a mini-implant of 6 mm should be sufficient,1 and the consequent higher insertion torque moment could even call for predrilling in order to decrease the risk of implant fracture or secondary implant failure.24,25
The relatively short distance to the maxillary sinus from the outer alveolar cortical bone found in the present study implies a risk of perforation of the maxillary sinus if mini-implants are planned for insertion with an apical inclination. A recent study10 measured the bone depth perpendicular to the bone surface and found a minimum distance from the palatal aspect, slightly apical to the mid-root level, to the sinus as low as 1.90 mm distally and 0.80 mm mesially to the first maxillary molar. Having in mind that these specific interradicular sites are recommended as safe sites for insertion of mini-implants,5,13,17 sinus perforations are expected to be a relatively common side effect. Nevertheless, not much attention has been given to this side effect, presumably because a small perforation rarely gives rise to complications and heals without intervention.26–28 The use of long mini-implants (8 mm) is recommended1 in order to increase primary stability when thin cortical bone (less than 1 mm) can be expected.1,2 In the maxillary molar region, insertion of 8-mm long mini-implants at 45° to the long axis of the teeth will increase the cortical bone-to-implant contact (compared to perpendicular insertion) and the trabecular bone-to-implant contact in cases with a sufficient distance to the sinus. Furthermore, the use of an 8-mm mini-implant will increase the chance of obtaining bicortical anchorage, shown by Brettin et al.29 to be superior to unicortical anchorage. However, apically angulated insertion in the maxillary molar region will lead to a higher risk of sinus perforation. The present study indicates that the risk of sinus perforation can be reduced without compromising the primary stability, if the 8-mm mini-implant is inserted as occlusally/marginally as possible, still with an apical direction.
A successful mini-implant insertion should lead to a stable implant without complications or injury to teeth or surrounding tissues. Hence, the use of skeletal anchorage should be applied only after careful diagnosis and planning of biomechanics in relation to the insertion site and vice versa.30
The interindividual variation was large with respect to the cortical bone thickness as well as the distance from the outer alveolar cortical bone to the maxillary sinus.
The thickness of the alveolar cortical bone is often less than 1 mm buccally and palatally in the entire maxilla, and buccally in the anterior mandible.
In the posterior mandible, the alveolar cortical bone is frequently thicker than 2 mm.
Changing the insertion angle from 90° to 45° can enhance primary mini-implant stability but will increase the risk of sinus perforation in the maxillary molar region.