The aim of this study was to evaluate the diagnostic potential of periapical radiograph, panoramic radiograph, and cone-beam computed tomography (CBCT) in detecting implant-related perforation of the inferior alveolar canals.
In this
For detection of pilot drill injuries by observer 1, CBCT (AUC = 1) and periapical radiograph (AUC = 0.889) were significantly better than using panoramic radiographs (AUC = 0.694) (
CBCT was better in detecting pilot drill injuries to the IAN canal compared to panoramic radiograph. Therefore, in cases where clinical presentations suggest IAN disturbances, CBCT scan should be preferred. However, the diagnostic potential of periapical radiograph, panoramic radiograph, and CBCT was not significantly different for detection of penetrative injuries to the IAN canal.
Dental implants are widely used for the replacement of missing teeth in modern dentistry.[
Radiographic examinations are performed at different stages of implant treatments, for treatment planning, fabrication of implant guides, and follow-up of inserted implants.[
Limited studies have been performed on the potential of different imaging techniques in the diagnosis of implant-related injuries to the IAN. In 2020, Sirin
This
A sample size of 15 implants in each experimental and control group was needed based on the following equation, considering an alpha of 0.05 and a power of 75%:
where
Fifteen fresh adult sheep hemimandibles were used in this study.
A total of 45 dental implants (Bionik, Nik Kasht Asia, Tehran, Iran) made of commercially pure titanium were inserted according to the manufacturer’s instructions. Two groups of implant-related injuries to the canal were simulated: (1) injury by the pilot drill (with a diameter of 2 mm) and (2) penetration of the implant tip into the IAN canal. Fifteen implants were inserted for each group using direct vision from the prepared window. For the pilot drill injury, the superior border of the IAN canal was penetrated using the pilot drill. However, the implant was placed 1 mm above the superior border of the canal. To simulate the penetration of the implant tip into the inferior alveolar canal, the implant tip was placed 1 mm into the canal. Fifteen implants were used for the control group, where the implant tip was placed 1 mm above the superior border of the nerve canal, and no pilot drill injury was performed during the drilling procedure.[
A cranium model (Anatokala, Tehran, Iran) with the maxilla and mandible in occlusion was prepared. The posterior segment of the mandible was excised with preservation of the inferior border of the mandible. The sectioned portion of the sheep hemimandibles with inserted implants was then located and secured in the prepared cranium model. Soft tissue was simulated using 15 mm of red wax (Polywax, Izmir, Turkey) to make the phantom ready for imaging.[
Digital periapical radiographs were obtained by an intraoral radiographic unit (Planmeca, Helsinki, Finland) with the parallel technique using size 2 intraoral imaging plates (Durr Dental, Bietigheim-Bissingen, Germany) and film-holders (Kerr, CA, USA) with exposure parameters of 63 kVp, 8 mA, and 0.1 s. The periapical images were viewed. The phantom was positioned for obtaining panoramic radiographs with the Frankfurt plane placed parallel to the horizontal plane and the mid-sagittal plane perpendicular to it. Panoramic radiography was obtained using the ProMax unit (Planmeca, Helsinki, Finland) with exposure parameters of 64 kVp, 5 mA, and 15.6 s. The same position of the skull phantom was replicated for obtaining CBCT images using a Galileos CBCT scanner (Sirona, Bensheim, Germany) with parameters of 85 kVp, 21 mAs, 280 μm voxel size, and field of view of 15 cm × 15 cm [
(a) Periapical radiographs of control implants, and (b) periapical radiographs of implants with pilot drill injury to the inferior alveolar nerve canal (right implant) and penetration of the inferior alveolar nerve canal (left implant). Arrows point to inferior alveolar nerve canal.
(a) Cropped panoramic radiographs of control implants, and (b) cropped panoramic radiographs of implants with pilot drill injury to the inferior alveolar nerve canal (right implant) and penetration of the inferior alveolar nerve canal (left implant). Arrows point to inferior alveolar nerve canal.
(a) Cross-sectional view of control implant, and (b) tangential cone-beam computed tomography views of implants with pilot drill injury to the inferior alveolar nerve canal (right implant) and penetration of the inferior alveolar nerve canal (left implant). Arrows point to inferior alveolar nerve canal.
Images were viewed in the following software: digital periapical radiographs in Scanora (Soredex, Tuusula, Finland), digital panoramic radiographs in Romexis (Planmeca, Helsinki, Finland), and CBCT images in Sidexis 4 (Sirona, Bensheim, Germany). Two observers (a radiologist with 5 years of experience and an oral surgeon with 10 years of experience) performed the interpretation of images in a a quiet room with dim lighting. The observers were blind to the category of each image, and all images were displayed randomly for the observers. The observers were free to use different software options such as contrast, brightness, sharpen, and zoom. In addition, in the CBCT images, the observers could view the images in any desired view. For evaluating each implant, the observers were asked to determine whether canal injury was present or not, and also the type of injury present, i.e., pilot drill injury or penetration of the implant tip into the canal. The observers’ responses were recorded using a 4-point Likert scale: (1) canal injury is absent, (2) pilot drill injury is present, (3) penetrative injury is present, and (4) uncertain. After 2 weeks, the observers were asked to evaluate 20% of the images once again.
Intraobserver and interobserver agreements were determined using Cohen’s kappa. Images with score 4 (uncertain) were removed from further analysis. Diagnostic accuracy was calculated for each imaging modality using the area under the receiver operating characteristic curve (AUC). Statistical analysis was performed using SPSS software (version 25, IBM, NY, USA). In addition, AUC values of different modalities were compared using MedCalc statistical software version 19.2.6 (MedCalc Software Ltd., Ostend, Belgium;
For all implant groups and imaging modalities, the intraobserver agreements calculated by kappa values ranged from 0.860 to 1.00, indicating strong to almost perfect agreements.
Interobserver agreements determined by kappa values were strong to almost perfect (0.806–0.900) for periapical radiographs, moderate to strong for panoramic radiographs (0.544–0.772), and strong to almost perfect for CBCT image sets (0.746–0.903).
For detection of intact superior border of the IAN canal below the control implants, AUC values for observer 1 were 0.917, 0.817, and 1 for periapical radiograph, panoramic radiograph, and CBCT, respectively. These values for observer 2 were 0.850, 0.767, and 0.917 for periapical radiograph, panoramic radiograph, and CBCT, respectively [
Area under the receiver operating characteristic curve for each observer in detecting intact roof canal below control implants
Pairwise comparison of the modalities showed that CBCT had a significantly higher AUC for detection of intact IAN canal roof below the control implants compared to panoramic radiography (
For detection of pilot drill injury to the IAN canal, AUC values for observer 1 were 0.889, 0.694, and 1 for periapical radiograph, panoramic radiograph, and CBCT, respectively. The corresponding values for observer 2 were 0.842, 0.683, and 0.897 [
Area under the receiver operating characteristic curve for each observer in detecting pilot drill injury
Pairwise comparison revealed that for detection of pilot drill injuries by observer 1, CBCT and periapical radiography were significantly better than panoramic radiography (
For detection of penetration of implant tip to the IAN canal, AUC values for observer 1 were 0.995, 0.990, and 1 for periapical radiography, panoramic radiography, and CBCT, respectively. The corresponding values for observer 2 were 0.986, 0.948, and 0.995 [
Area under curve (AUC) for each observer in detecting penetration injury
Pairwise comparison for AUC of different modalities for detection of penetrative injuries to the IAN canal revealed that for observers 1 and 2, no significant difference was noted between the three modalities in detecting penetration of implants to the IAN canal (
Based on our findings, CBCT performed better than panoramic radiographs in visualizing pilot drill injuries to the IAN canal. However, for detecting penetration of implants into the IAN canal, the discriminatory performance of periapical, panoramic radiographs, and CBCT imaging was not significantly different.
Renton
In this study, periapical, panoramic, and CBCT images were used to evaluate the presence or absence of damage to the superior border of the IAN canal by dental implants and drills. These imaging modalities are the most common techniques used in different stages of treatment with dental implants, and evaluating their accuracy in these phases is valuable for determining their role in implantology. A panoramic radiograph is a combination of scanning and tomography and is therefore prone to distortion and superimposition, as well as loss of resolution for structures located outside the imaging layer or focal trough. In addition, linear measurements in panoramic radiographs are not accurate. Due to these shortcomings, the application of panoramic radiographs in implantology is better to be reserved only for when an overview of the implants and adjacent dentoalveolar structures is required. Periapical radiography can provide high-resolution images from implants and dentition with minimal radiation dose. Periapical radiographs obtained by the parallel technique using film holders can provide a view of the dentition with minimum distortion. However, they still have the disadvantages of 2D images, including superimposition. Neto
Several studies have focused on the safety margin when discussing the proximity of dental implants to the IAN canal in different images. Basa and Dilek in their study evaluated the density and thickness of the superior border of the IAN canal in CBCT images. According to their findings, the cortical density of the canal borders has to be investigated prior to implant placement, as canals with thin walls cannot withstand the pressures of implant drilling and may collapse during implant insertion just above the canal roof.[
In the present study, two of the most common types of injury to the IAN were simulated. In the penetrative injury to the IAN canal, the implant tip was located in the canal. Whereas, in the pilot drill injury, although the superior border of the canal was perforated by the implant drills, the final position of the implant remained above the canal outline. Based on the injury patterns simulated in the present study, neurosensory disturbances observed despite an apparent distance between the implant and the inferior alveolar nerve may be attributed to pilot drill–related injury or the transmission of occlusal forces to the canal. CBCT provides 3D visualization of structures and allows for better detection of localized perforations of the IAN canal’s roof caused by pilot drill injuries compared with the 2D view of periapical and panoramic radiographs. While in penetrative injuries, with the implant tip completely in the IAN canal, panoramic radiography provided similar diagnostic potential as compared with periapical radiography and CBCT imaging.
To the authors’ knowledge, only one previous study has been performed on the simulation of different types of injuries to the IAN and that has compared the diagnostic ability of different imaging modalities. Sirin
In 2019, Vanderstuyft
One of the limitations of this study is the potential difference between the bone density and shape and size of the IAN canal of sheep and human mandibles which can limit the generalization of the findings. In addition, using implants with different material and designs can alter the present results. Moreover, the diameter of the pilot drill in the implant drilling system determines the dimension of the pilot drill injury, making it easier or more difficult to detect. Further studies on these subjects are recommended with application of artifact reduction algorithms in CBCT images. In addition, performing similar studies using different implant materials such as zirconia is also suggested.
CBCT was better in detecting pilot drill injuries to the IAN canal compared to panoramic radiograph. Therefore, in cases where clinical presentations suggest IAN disturbances, CBCT scan should be preferred. However, the diagnostic potential of periapical radiograph, panoramic radiograph, and CBCT were not significantly different for detection of penetrative injuries to the IAN canal.
This study was financially supported by Isfahan University of Medical Sciences (#1400154).
The authors of this manuscript declare that they have no conflicts of interest, real or perceived, financial or nonfinancial in this article.
