Evaluation of vertical magnification coefficients of potential dental implant site in cone-beam computed tomography images: An ex vivo study

Bilgun Cetin; Faruk Akgunlu

doi:10.4103/srmjrds.srmjrds_129_20

ORIGINAL ARTICLE

Year : 2021 | Volume : 12 | Issue : 2 | Page : 61-66

Evaluation of vertical magnification coefficients of potential dental implant site in cone-beam computed tomography images: An ex vivo study

Bilgun Cetin¹, Faruk Akgunlu²
¹ Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Burdur Mehmet Akif Ersoy University, Burdur, Turkey
² Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Selcuk University, Konya, Turkey

Date of Submission	02-Dec-2020
Date of Decision	13-Jan-2021
Date of Acceptance	02-Mar-2021
Date of Web Publication	30-Jun-2021

Correspondence Address:
Dr. Bilgun Cetin
Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Burdur Mehmet Akif Ersoy University, Istiklal Yerleskesi, Burdur 15030
Turkey

Source of Support: None, Conflict of Interest: None

Check

DOI: 10.4103/srmjrds.srmjrds_129_20

Abstract

Purpose: This study aimed to investigate whether the vertical magnification coefficient (MC) changes with errors in head positioning in cone-beam computed tomography (CBCT), which is increasingly used for measurements in implantology. Materials and Methods: CBCT images were obtained in 15 different positions (5° in one or two planes) with the help of a positioner that allowed angular movement in coronal, sagittal, and horizontal planes, using a dry skull, in which three steel spheres were placed in the alveolar bone of each tooth region. An observer measured the vertical dimensions of the spheres in the images twice. Vertical MCs were calculated by dividing the measurements by the real diameter lengths of the spheres. The inter-class correlation coefficient (ICC) was used to compare the measurements repeated twice. Moreover, the Kolmogorov–Smirnov, Levene, one-way analysis of variance, Kruskal–Wallis and least significant difference, Tukey, and Tammhane binary comparison tests were performed in the statistical analysis. Results: A total of 1260 measurement values were used. There was high reliability between the first and second measurement values (ICC = 0.986). The measurements in the images taken in the ideal position were quite close to the physical size. A statistically significant difference was found between the ideal position and the images taken in other positions. The most affected regions by the different positions were the anterior regions for both jaws. Conclusions: Vertical measurements in CBCT images are reliable; however, especially in the anterior regions, such measurements can vary significantly with errors in positioning of patients' heads.

Keywords: Cone beam computed tomography, dental implants, magnification coefficient, vertical dimension

How to cite this article:
Cetin B, Akgunlu F. Evaluation of vertical magnification coefficients of potential dental implant site in cone-beam computed tomography images: An ex vivo study. SRM J Res Dent Sci 2021;12:61-6

How to cite this URL:
Cetin B, Akgunlu F. Evaluation of vertical magnification coefficients of potential dental implant site in cone-beam computed tomography images: An ex vivo study. SRM J Res Dent Sci [serial online] 2021 [cited 2023 May 28];12:61-6. Available from: https://www.srmjrds.in/text.asp?2021/12/2/61/319861

Introduction

Dental implants used in the place of missing teeth are considered to be one of the most important development in dental science in the past 50 years.^[1],[2] The development of modern osseointegration methods and its high success rate have increased replace missing teeth with implants.^[2] For success, it is highly important to make the correct measurements before the implant surgery.^[3] Although various imaging methods are used, cone-beam computed tomography (CBCT) images provide the closest to true measurements. These images show the alveolar bone density, height, and buccolingual width at any point of the jawbones, as well as accurately showing pathologies, bone inclination, and vital anatomical structures.^[4]

Although CBCT provides high-resolution and dimensionally accurate images and measurements of potential implant areas, reconciling areas measured by different observers and at different time points remain challenging. Studies have reported conflicting results in terms of measurement accuracy in CBCT scans.^[5] A review of the literature showed some studies reporting that the accuracy of the measurements on the CBCT images can change due to incorrect positioning of the jaws during patient positioning; in contrast, other research suggests that the measurements resulting from the incorrect positioning do not affect the accuracy of the CBCT.^{[6],[7],[8],[9],[10]} Moreover, in some recent studies, it has been reported that linear lengths measured on CBCT images are shorter than the true length.^{[9],[11],[12],[13]}

Based on previous studies with confusing results, it is aimed to evaluate the reliability of measuring short distances of potential implant regions in CBCT images. And also, it was aimed to determine how much the bone height measured in patients with reduced bone tissue changes when there are deviations in head position. For this, three small-sized spheres were used for each tooth region, and images were obtained within both one and two planes deviation in the head position.

Materials and Methods

First, a dry skull with an orthodontically correct tooth sequence was obtained from the Selçuk University, Faculty of Medicine, Anatomy Department. Twenty-eight tooth regions in the dry skull were brought to a sufficient length and width. A micromotor and a bevel-shaped burr-drill were used. Then, three metal spheres (3.15 or 2 mm diameter) were prepared for each tooth region (wisdom teeth were excluded); these were placed in the slots at equal intervals (1.77 mm) and fixed with the help of modeling wax. A total of 84 (3 × 28) stainless steel spheres were used after checking their diameters with the help of digital calipers. Since the lower incisors did not have enough thickness of bone, 2 mm diameter spheres were placed in only this region. Spheres with a diameter of 3.15 mm were placed in the remaining regions. These small diameters were used to correspond to precise measurements, especially in patients with small amounts of alveolar bone remaining. The dry skull was fixed by adhering it to a tripod positioner that permits angular position change in coronal, sagittal, and horizontal planes by means of a cylindrical part that passes through the foramen magnum [Figure 1].

Figure 1: The positioner that permits angular position change in all three planes and the dry skull prepared to cone-beam computed tomography scanning

Click here to view

Second, 15 CBCT images for both jaws were obtained using head position errors. The coronal plane was expressed as “X,” and in this plane, values of “+” for the right tilt of the head and “−” for the left tilt were used. The sagittal plane was expressed as “Y,” and in this plane, a value of “+”was assigned to flexion (forward-tilt) and “−” to extension (backward-tilt). As for the horizontal plane, this was expressed as “Z,” and in this plane, a value of “+” was used for the rotation to the right and “−” to the left. After trial images are taken, some region of the jaws did not appear in the images taken tilt or rotation more than 5° and since there is no significant difference in the images taken with 1° increments, it was decided to use 5° angles in a maximum of two planes [Table 1].

Table 1: Fifteen head positions that were determined in three planes

Click here to view

Images using a Kodak 9000/3D system, Carestream Health Inc., Rochester, NY, USA, were taken with an average of 35 s of scanning time for the lower jaw and an average of 32 s of scanning time for the upper jaw using values of 70 kV and 10 mA. Vertical lengths of the spheres in all the images obtained were measured twice by a single observer using the same monitor (Dell Inspiron 15, 1366 × 768 resolution, TX, USA). Measurements were made in the cross-sectional section (1 mm thick) and by crossing the middle of the spheres [Figure 2].

Figure 2: (a) Cone-beam computed tomography measurement screen (the vertical dimension measurement was performed on cross-sectional sections passing through the middle in the spheres.). (b) Cone-beam computed tomography images show at 5° extension and left tilt (0, -5, -5) image of cross sections from the upper and the lower jaw

Click here to view

Finally, vertical magnification coefficient (MC) was calculated by dividing the measurements by the actual diameter lengths and statistical analyses were made using the results. The jaws are divided into six regions in order to increase the reliability of the statistical results and to reduce the error margin. The regions are lower right posterior (right molar and premolar teeth), lower anterior (canine and incisors), lower left posterior (left molar and premolar teeth), upper left posterior (left molar and premolar teeth), upper anterior (canine and incisors), and the upper right posterior (right molar and premolar teeth). While investigating how positioning affects vertical MC between the alveolar bone levels (occlusal, middle, and apical), jaws are divided into four regions to increase statistical reliability. The regions are lower posteriors (all lower molar and premolar teeth), lower anterior (canines and incisors), upper posteriors (all molar and premolar teeth), and upper anterior (canines and incisors). However, in case of the jaws not positioned symmetrically (for example, right or left rotation) would not be correct to evaluate the posterior regions together. Hence, the difference between the alveolar bone levels was evaluated only in the positions showing right–left symmetry ([0, 0, 0], [0, +5, 0], [0, −5, 0]). While creating the regions, it was taken into consideration that there was no difference between the right and left posterior regions in the ideal position.

Statistical analysis

Statistical analyses were made with the IBM SPSS software program (version 21, Spss Inc., Chicago, IL, USA). The interclass correlation coefficient (ICC) was used to compare the measurements repeated twice. The second measurements were used for analysis since there was high reliability between the first and second measurement values (ICC = 0.986). The assumption of conformity to normal distribution was analyzed with the Kolmogorov–Smirnov test and the homogeneity assumption of the variances was analyzed with the Levene test. In cases, parametric test assumptions were provided, the one-way analysis of variance test was used for more than two groups with one variable. In cases, parametric test assumptions are not provided, the Kruskal–Wallis test was used for more than two groups. Then, least significant difference, Tukey, Tammhane binary comparison tests were performed. Five percent significance level was used.

Results

In the study, a total of 1260 (28 × 15 × 3) measurements were used in each of the images obtained with 15 different positions.

Effects of positioning on vertical magnification coefficient of different regions

All average vertical MC values obtained from the regions were all >”1” proving the CBCT device used shows slightly increased values than the actual length [Table 2] and [Table 3].

Table 2: Kruskal-Wallis test results within different regions of the lower jaw

Click here to view

Table 3: Kruskal-Wallis test results within different regions of the upper jaw

Click here to view

It was found that MCs in the right posterior region of the lower jaw increased statistically significantly in (0, +5, +5), (+5, 0, 0), (+5, 0, −5), and (+5, 0, +5) positions compared to the ideal one. The region was not affected by the movement in the Y plane, especially the extension. However, it was observed that was more affected by head movements to right. On the other hand, the left posterior region had statistically significant in (0, +5, +5) and (0, −5, −5) positions. It shows that the right and left regions of the jaws are not symmetrically affected by the head position errors. In anterior region of the lower jaw, it was found that MCs grew significantly in (0, 0, −5), (0, +5, 0), (0, +5, +5), (−5, 0, 0), (−5, 0, −5), (−5, 0, +5), (+5, 0, 0), (+5, 0, −5), and (+5, 0, +5) positions. Thus, in the lower jaw, the anterior region was the most affected by the change of position.

It was found that MCs in the right posterior region of the upper jaw increased statistically significantly in (0, 0, +5), (0, −5, 0), and (0, −5, +5) positions compared to the ideal one. The region was not affected by the movement of the head in the X plane. Similarly, the left posterior region was not affected by movements in the X plane. The left region was affected in (0, 0, −5), (0, −5, 0), (0, −5, −5) positions. The most affected area was again identified as the anterior region. It was affected in (0, 0, +5), (0, 0, −5), (0, +5, 0), (0, −5, 0), (0, +5, +5), (0, −5, −5), (0, −5, +5), (+5, 0, +5) positions. However, the anterior region of the upper jaw was less affected by the movement of the head in the X plane. Therefore, MCs in the upper jaw did not undergo statistically significant changes with the tilting to the left or right.

Effects of positioning on vertical magnification coefficient of different alveolar bone levels

Average vertical MCs and statistic results obtained for the alveolar bone levels are shown in [Table 4]. There was no statistically significant difference in MCs of both jaws between alveolar bone levels in the CBCT image taken with the ideal position. However, in the image taken with (0, +5, 0), (0, −5, 0) positions, in the posterior teeth region, MCs of the apical level were larger than the occlusal level in the lower jaw (P = 0.027, P = 0.017). MCs at the levels of the alveolar bone in the posterior of the upper jaw were not affected by extension or flexion. However, in the anterior of the upper jaw, there was a significant difference at (0, −5, 0) position (P = 0.010), which apical level was larger than the occlusal level.

Table 4: One-way ANOVA test results within different alveolar bone levels

Click here to view

Discussion

Examining radiological images of the jaws are vital in evaluation before implant surgery. While choosing an implant length that leaves a safety gap of at least 2 mm between the tip of the implant and important anatomical structures, making correct measurements is imperative to prevent possible complications.^[14] However, errors in head positioning may cause distortions in images and inaccuracies in measurements. This may result from incorrect implant size selection.^[8] Therefore, the current study was aimed to investigate whether the vertical size changes with errors in head positioning in CBCT images.

Size measurements in CBCT were found more accurate than all other imaging methods that are very often used for implant surgery.^[3] The American Academy of Oral and Maxillofacial Radiology recently reported that CBCT was the best option for implant planning.^[15] Although CBCT offers high resolution and dimensionally true images, accurate measurements measured by different observers and at different time points remain challenging. Studies have reported contradictory results regarding measurement accuracy in CBCT scans.^[5] Luangchana et al. used different dry skulls and field of views (FOV) found that the linear measurements were less than their physical measurements.^[9] Similarly, a few studies reported that there were some smaller measurements in CBCT than the actual size.^{[11],[12],[13]} Nevertheless, in the present study, the mean vertical MC was found 1.09 ± 0.031 in the image taken with the ideal position. This means it was measured slightly larger than the physical size. This may be due to device differences.

Alternatively, in a study of the reliability of CBCT measurements by Kamburoğlu et al., gutta-percha was attached to various points of a skull. The distance between them was measured both physically on the skull and in various reconstructions on images from two different CBCT devices.^[10] In the study, they concluded that the measurements were quite close to physical measurements.^[10] For similar purposes, Cook et al. conducted a study comparing the height and width of alveolar bone with direct measurements obtained by dissection from cadaver with three different FOV and voxel scan options. As a result, they reported CBCT images are reliable for both height and width, even if their FOV and voxel dimensions are different.^[16] In this study, a single FOV and a single device were used, and compatible with the studies, measurements in the ideal position were very close to the actual. In some other studies likewise, it has been reported that the linear measurements are very close to the actual length.^{[7],[11],[17]} This situation indicates that, even if values are slightly below or slightly above, it is safe to make linear measurements on CBCT images taken from a head positioned following the manufacturer's instructions. However, the purpose of this study was to determine whether the measurements had changed significantly if patient's head deviated from the ideal position.

Icoz and Akgunlu reported that, when CBCT images were taken in the ideal position following the manufacturer's instructions, values that would be very close to real size, however, changes in the head position cause a significant growth in the vertical dimension.^[18] The results of the present study are consistent. On the other hand, even in the ideal position, a little magnification was found in this study. These findings may be associated with the size of the spheres which were much smaller than the pins in the other study, so there was a need for more sensitive measurement. That emphasizes that measurement should be performed more carefully in patients with bone loss.

Sabban et al. made horizontal and vertical measurements in CBCT scans using different head positioning angles and compared them with the ideal position. They reported that statistical analysis of the vertical measurements, average errors ranging from − 2 mm to 3 mm in various head positions, was mainly seen in the extension of the head position and in posterior regions.^[5] This resulting size difference was thought to be due to the angles used (15°–20°) and the measured distance was larger than this study. In addition, in the present study, since the angular positioning was performed simultaneously in two planes, the dimensional deviation was mostly observed in the anterior regions. However, as a result of the study of Sabban et al., the angular errors in head positioning changed the linear measurements significantly as in the present study.^[5]

In this study, vertical MC in the anterior regions changed significantly with different head position errors in all three planes, but the MCs in the upper posterior regions differed in several positions simultaneously in two planes. Adibi et al. found that the average measurement error was significantly higher for the right or left tilt position compared to the ideal position,^[6] although Sheikhi et al. showed that rotation and extension positions were the deviations that have the greatest impact on the accuracy of CBCT measurements.^[8]

There are certain limitations to this study. CBCT images with deviations of more than 5° could not be examined, as some regions were not visible in the image. Furthermore, since steel material was used, metal artifact can affect the reliability of the measurements by preventing the borders of the spheres from appearing sharp.

Conclusions

The vertical measurements in the CBCT images taken with the ideal position are reliable. Yet when there is an error in head position, variable results are obtained depending on the CBCT device used and the extent of the position error. Therefore, training staff to take CBCT scans is extremely important in preventing possible complications. In short lengths, additional attention should be paid to image scanning and measurements should be done more sensitively. When positioning error is detected in an evaluated image, it is recommended to measure it by adding 10%–20% of confidence interval (according to vertical MCs in this study).

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Angelopoulos C, Aghaloo T. Imaging technology in implant diagnosis. Dent Clin North Am 2011;55:141-58.
2.	Buser D, Sennerby L, De Bruyn H. Modern implant dentistry based on osseointegration: 50 years of progress, current trends and open questions. Periodontol 2000 2017;73:7-21.
3.	Yim JH, Ryu DM, Lee BS, Kwon YD. Analysis of digitalized panorama and cone beam computed tomographic image distortion for the diagnosis of dental implant surgery. J Craniofac Surg 2011;22:669-73.
4.	Egbert N, Cagna DR, Ahuja S, Wicks RA. Accuracy and reliability of stitched cone-beam computed tomography images. Imaging Sci Dent 2015;45:41-7.
5.	Sabban H, Mahdian M, Dhingra A, Lurie AG. Tadinada A. Evaluation of linear measurements of implant sites based on head orientation during acquisition: An ex vivo study using cone-beam computed tomography. Imaging Sci Dent 2015;45:73-80.
6.	Adibi S, Shahidi S, Nikanjam S, Paknahad M, Ranjbar M. Influence of head position on the CBCT accuracy in assessment of the proximity of the root apices to the inferior alveolar canal. J Dent Shiraz Univ Med Sci 2017;18:181-6.
7.	Ganguly R, Ruprecht A, Vincent S, Hellstein J, Timmons S, Qian F. Accuracy of linear measurement in the Galileos cone beam computed tomography under simulated clinical conditions. Dentomaxillofac Radiol 2011;40:299-305.
8.	Sheikhi M, Ghorbanizadeh S, Abdinian M, Goroohi H, Badrian H. Accuracy of linear measurements of galileos cone beam computed tomography in normal and different head positions. Int J Dent 2012;2012:214954.
9.	Luangchana P, Pornprasertsuk-Damrongsri S, Kiattavorncharoen S, Jirajariyavej B. Accuracy of linear measurements using cone beam computed tomography and panoramic radiography in dental implant treatment planning. Int J Oral Maxillofac Implants 2015;30:1287-94.
10.	Kamburoğlu K, Kolsuz E, Kurt H, Kılıç C, Özen T, Paksoy CS. Accuracy of CBCT measurements of a human skull. J Digit Imaging 2011;24:787-93.
11.	Stratemann SA, Huang JC, Maki K, Miller AJ, Hatcher DC. Comparisonof cone beam computed tomography imaging with physicalmeasures. Dentomaxillofac Radiol 2008;37:80-93.
12.	Lascala CA, Panella J, Marques MM. Analysis of the accuracy of linear measurements obtained by cone beam computed tomography (CBCT-NewTom). Dentomaxillofac Radiol 2004;33:291-4.
13.	Baumgaertel S, Palomo JM, Palomo L, Hans MG. Reliability andaccuracy of cone-beam computed tomography dental measurements. Am J Orthod Dentofacial Orthop 2009;136:19-25.
14.	Vazquez L, Nizam AL, Din Y, Christoph-Belser U, Combescure C, Bernard JP. Reliability of the vertical magnification factor on panoramic radiographs: Clinical implications for posterior mandibular implants. Clin Oral Implants Res 2011;22:1420-5.
15.	Pedroso LA, Garcia RR, Leles JL, Leles CR, Silva MA. Impact of cone-beam computed tomography on implant planning and on prediction of implant size. Braz Oral Res 2014;28:46-53.
16.	Cook VC, Timock AM, Crowe JJ, Wang M, Covell DA Jr. Accuracy of alveolar bone measurements from cone beam computed tomography acquired using varying settings. Orthod Craniofac Res 2015;18 Suppl 1:127-36.
17.	Araki K, Maki K, Seki K, Sakamaki K, Harata Y, Sakaino R, et al. Characteristics of a newly developed dentomaxillofacial X-ray cone beam CT scanner (CB MercuRay): System configuration and physical properties. Dentomaxillofac Radiol 2004;33:51-9.
18.	Icoz D, Akgunlu F. Effects of positioning upon the vertical dimension on cone beam computed tomography. Edorium J Dent 2016;3:40-4.