Assessment of cone beam computed tomography for determining position and prognosis of interradicular mini-implants

ABSTRACT Objective: To investigate the influence of dynamic visualization of cone beam computed tomography (CBCT) scans on orthodontist’s assessment of positioning status and prognosis of interradicular mini-implants (MI). Methods: Three MI positions were virtually simulated in thirty CBCT volumes: (1) MI 1 mm from the lamina dura (LD), (2) MI touching the LD and (3) MI overlapping the LD. Each position was exposed to orthodontists (n = 35) as panoramic reconstruction, sagittal reconstruction and a sequence of axial slices. Each orthodontist evaluated the MI position (relationship with the LD) and scored the prognosis using a four-point scale (the higher the score, the better the prognosis). Kappa, Friedman and Nemenyi statistics were used. Results: Statistically significant associations were detected between the prognosis scores and the type of image visualized (p<0.05). The dynamic visualization of the CBCT volume (axial slices) was associated with higher scores for prognosis and more reliable evaluation of MI positioning. Inconsistent outcomes were more frequently associated with panoramic and sagittal reconstructions. Conclusion: The dynamic visualization of axial slices allowed orthodontists to perform better assessment of MI position and considerably affected prognosis judgment.


INTRODUCTION
In Orthodontics, specific movements that were challenging in the past, such as distalization and intrusion, became more feasible with the advent of mini-implants (MI). [1][2][3] In addition to proper biomechanics, the MI reduce anchorage loss. 4 Clinical risks, however, may occur especially when MI damage the periodontal ligament or adjacent roots 5,6 and trigger external root resorption. 7 In order to minimize root resorption and promote proper bone anchorage, MI should be placed at least 1 mm far from the lamina dura. 8 From the perspective of treatment prognosis, having MI in contact with root and periodontal ligament is one of the main causes of MI failure. 6,[9][10][11][12][13] Studies with preoperative radiographs were designed to find out safer anatomic regions for MI placement (i.e. regions with more interradicular space), 14,15 but the available space varies between patients due to the broad spectrum of skeletal discrepancies, axial tooth angles and anatomic variations. 14 Postoperative imaging, on the other hand, might be straight to the point when it comes to the assessment of dentoalveolar damages. 16 Protocols that include postoperative radiographs, however, may alter clinical judgment about the final position of the MI -without necessarily, increasing certainty. 17 Batista Junior ES, Franco A, Soares MQS, Nascimento MCC, Junqueira JLC, Oenning AC -Assessment of cone beam computed tomography for determining position and prognosis of interradicular mini-implants 5 Image superimposition hinders a clear visualization of dentomaxillofacial structures in two-dimensional images. 11,15,18 Differently, three-dimensional imaging, namely cone beam computed tomography (CBCT), improves assessment of the relationship between MI and adjacent teeth and bone. 4,16 Higher radiation dose and cost restrict the use of CBCT 19

ETHICAL ASPECTS AND STUDY DESIGN
This cross-sectional observational study was carried out with the approval of the institutional committee of ethics in human research (protocol: 3.651.240).

SAMPLE
Thirty CBCT volumes, stored in DICOM format, were selected from the image database of a public university. All the images were acquired in an i-CAT ® device (Imaging Sciences International Inc., Hatfield, PA, USA). The CBCT scans were acquired with a field of view (FOV) of 23 cm x 17 cm, given that this FOV size is frequently used for orthodontic indications, and consequently, it is also used for mini-implant planning.
The inclusion criteria consisted of CBCT volumes of males and females taken with the same acquisition and reconstruction settings, to avoid the inclusion of bias due to differences in noise level and/or spatial resolution. The patients included should be eligible for orthodontic treatment with MI and should have maxillary first molars and second premolars, as well as sound bone structure in the region. Patients under 18 years of age, with metallic restorations or prosthetic materials in the region or any type of anatomic variation or lesion in the adjacent alveolar bone and maxilla were excluded.
The simulation of surgical placement of MI was designed between the maxillary first molar and second premolar, within a distance of 2 mm from the bone crest and perpendicular to the adjacent teeth. 20

IMAGE ANALYSIS
Thirty-five orthodontists were invited to perform image analysis (mean age of 34 years, SD 5.08, range 27-43 years).
The inclusion criteria for the examiners consisted of previous experience with history of orthodontic practice and knowledge of MI therapeutics for at least three years. The set of 10 Kappa statistics assessed intra-examiner reproducibility.
Data analysis was performed with R software (R foundation, Vienna, Austria) with significance level of 5%. Analyzing the scores (1 -far from the LD, 2 -touching the LD, 3 -overlapping the LD, 4 -impossible to determine) from another perspective, when "impossible to determine" (score 4) was considered during data analysis, lack of difference was found between image types (Table 2). Thus, the score data were analyzed by removing the score 4. In that sense, the higher the score, the higher would be the proximity to the LD. As a result, statistically significant differences were detected between the image types for MI "overlapping the LD" (     Comparisons based on prognosis scores are found in Table 5 (the higher the score, the better the prognosis). Compared with reconstructions, stronger and statistically significant associations were found when the MI was positioned far from the LD / overlapping the LD and observed through CBCT volumes (p<0.05).

ASSESSMENT OF MI POSITIONING
In other words, only in CBCT volumes there was a statistical distinction of the prognosis among the three MI positions.   Table 4: Agreement between examiner's scores for mini-implant prognosis and their position in the three image types.  Apparently, CBCT analyses might not be necessary for installing MI with anchorage in the palate (close to the median maxillary suture). 25 Lateral radiographs may enable correct and reliable assessment of bone thickness prior to surgery, while CBCT would be of major value in complex cases with borderline anatomic features. 26,27 This is why the sample collected in the present study consisted of MI anchored in the alveolar process.