Evaluation of artifacts in cone-beam computed tomography from different intraradicular pins and cementation materials using different exposure protocols

Abstract

Purpose

To evaluate the generation of artifacts from intraradicular pins and cementation materials in cone-beam computed tomography images, using different exposure protocols.

Materials and Methods

Sixty extracted single-rooted premolars human were divided into six groups such as G1: Fiberglass pin with Allcem cement, G2: Metal core fiberglass pin core with Allcem cement, G3: metal core fiberglass pin with Allcem Core cement, G4: Fiberglass pin with Allcem Core cement, G5: Fiberglass pin with Panavia cement, and G6: metal core fiberglass pin with Panavia cement. CBCT images were acquired on the Orthophos XG^® 3D (Sirona Dental Systems, Bensheim, Germany) following the protocols with the exposure parameters; 1) high definition (HD) mode (70 kV and 6 mA, voxel size 100 µm, scan time 14s, 500 basis images), and 2) non-HD mode (70 kV and 10 mA, voxel size 160 µm, scan time 5s, 200 basis images). Two-way analysis of variance (ANOVA) and Tukey HSD post-hoc tests, Fisher’s exact test and the McNemar test were performed to detect statistically significant factors.

Results

For both HD and non-HD acquisition modes, the percentage of severe artifacts was higher in groups G3 and G6 (P<0.05). In the quantitative analysis, no significant differences were found among the parameters studied after accounting for all variables (P>0.05).

Conclusion

The type of fiberglass pin and cement used did not affect the production of the artifacts. When evaluating the scanning modes, the non-HD mode, which had lower mAs and fewer image bases, showed less artifact severity on the assessed CBCT scans.

Introduction

Cone-beam computed tomography (CBCT) allows the 3-dimensional evaluation of the structures of the dentomaxillofacial complex.1,2 Its application has been explored in clinical practice through diagnostic aid and surgical planning. In addition, several studies have investigated how to reduce the limitations of CBCT.3,4,5,6,7,8,9

One of these limitations is the generation of artifacts in the presence of high-density materials, which is a main factor that impairs the quality of diagnosis in CBCT.4 Artifacts are images that are inconsistent between the visual reconstruction and the actual content scanned.2,10 The formation of these artifacts occurs due to differences in the attenuation and absorption of X-ray beams by high-density materials, such as metal pins,11 gutta-percha, orthodontic appliances12 and dental implants.13,14,15

One way to reduce the generation of artifacts is to replace metallic materials (e.g., pins with metallic filaments or molten metallic cores) with other alternatives, such as fiber pins.16 Fiberglass pins are composed of a high mechanical strength epoxy resin that provides retention for the restorative material. However, some fiberglass pins may feature an internal metallic thread to increase their mechanical resistance against fractures, and with this, it can also increase the generation of artifacts.8

Several studies have investigated the influence of intra-radicular materials on the generation of artifacts;4,7,9,11,17 however, the studies cited did not consider the materials used for cementing intraradicular retainers. Since these cements may also influence the generation of artifacts. Furthermore, developing exposure protocols to minimise artifacts caused by intraradicular pins would help maintain CBCT specificity and enhance diagnostic accuracy.

In this context, this research aimed to evaluate the generation of artifacts from intraradicular pins and cementation materials in CBCT images under different exposure protocols.

Materials and Methods

This study protocol was approved by the Ethics and Research Committee (no. 61412916.9.0000.5181) and follows the Helsinki Declaration.

Sample preparation

Sixty extracted human teeth, specifically single-rooted premolars, were used for therapeutic purposes, and the sample was divided into six groups. All teeth were inspected by transillumination for the absence of root fractures, and radiographs were taken using phosphor storage plates to exclude teeth with pulp nodules, internal and/or external root resorption, previous endodontic treatment, multiple root canals, root canal obliteration, or any other anomalies.

After the cleaning and disinfection protocols, the crowns of the teeth were removed at the cementoenamel junction, and the root canals were standardized using the Reciproc R50 system (Dental Silver Reciproc - VDW, Munich, Germany) with thermomechanical compression filling. After 15 days, the roots were then prepared to two-thirds of their length (Peeso Long Drill nos. 1 and 2, Dentsply/Maillefer, São Paulo, SP, Brazil) for the installation of intraradicular pins.

Next, the teeth were randomly assigned to 6 groups (n = 10 per group) for the installation of intracanal materials such as G1: Fiberglass pin (Whitepost; Angelus, Londrina, PR, Brazil) with Allcem cement (FGM, Joinville, SC, Brazil), G2: fiberglass pin reinforced with metal core (Reforpost^®, Angelus, Londrina, PR, Brazil) with Allcem cement (FGM, Joinville, SC, Brazil), G3: fiberglass pin reinforced with metal core (Reforpost^®, Angelus, Londrina, PR, Brazil) with Allcem Core cement (FGM, Joinville, SC, Brazil), G4: Fiberglass pin (Whitepost; Angelus, Londrina, PR, Brazil) with Allcem Core cement (FGM, Joinville, SC, Brazil), G5: Fiberglass pin (Whitepost; Angelus, Londrina, PR, Brazil) with Panavia cement (Kuraray, Osaka, Japan), and G6: fiberglass pin reinforced with metal core (Reforpost^®, Angelus, Londrina, PR, Brazil) with Panavia cement (Kuraray, Osaka, Japan). All tooth groups have been prepared and cemented according to the manufacturer's recommendations.

Image acquisition

Each tooth was coated with a 0.2 mm layer of wax and placed individually in the alveoli of a lower right central incisor of a human mandible. The jaw was coated with a 5-mm-thick wax layer and immersed in a foam box filled with water to simulate the soft-tissue interference in the image.

CBCT images were acquired on the Orthophos XG^® 3D (Sirona Dental Systems, Bensheim, Germany) following the protocols with the exposure parameters pre-set modes provided by the imaging system itself: 1) high definition (HD) mode: 70 kV and 6 mA, voxel size 100 µm, scan time 14s, number of basis images 500; 2) non-HD mode: 70 kV and 10 mA, voxel size 160 µm, scan time 5s, number of basis images 200. The acquisitions were standardized with a field of view (FOV) of 8 cm Ø × 8 cm height. The device used the Feldkamp-Davis-Kress (FDK) image reconstruction algorithm. These volumes were exported in Digital Imaging and Communications in Medicine (DICOM) format.

Image preparation

Each DICOM file was read using the scanner’s native software (SIDEXIS XG, Sirona Dental Systems, Bensheim, Germany) and imported into ImageJ ver. 1.48 (National Institutes of Health, Bethesda, MD, USA, http://rsb.info.nih.gov/ij/).

To determine the region of interest (ROI), the total root length was measured, and the ROI was set exactly at the root midpoint (e.g., root length = 14mm; region of interest = 7 mm). Then, in an axial view, the resultant ROI images were set at an 8-bit scale (256 gray values) and saved with a black background.

Image analysis

Two examiners specialized in Dental Radiology analyzed the CBCT images independently. These evaluators, previously calibrated (Kappa = 0.81), were unaware of the procedures, methods, and conditions of the tooth roots. Observers were free to use the available imaging tools' options (density, contrast, and magnification). All images were evaluated under low light conditions, on a computer (Core I7, Intel^®, Santa Clara, California), with a 23.6″ HD 1920×1.80-pixel LED monitor, flat screen.

In the qualitative analysis, evaluators individually assessed the image artifact patterns and assigned scores for the degree of artifact interference in the CBCT images. 0 Absent (without the formation of artifacts), 1 Light (perceptible artifact), 2 Moderate (moderate amount of artifact) and 3 Severe (many artifacts generated) (Fig. 1).

Fig. 1. Classification of artifact severity according to qualitative analysis. A. Absent artifact. B. Light artifact. C. Moderate artifact. D. Severe artifact.

Initially, the "Freehand Selection" tool of ImageJ was used to select the area corresponding to the intraradicular materials (pin and cement), followed by the command "Edite > Clear" to erase the selected region, and the region was filled by the background image in black. Thus, only the region corresponding to the root dentin remains in the image (Fig. 2). In the case of a very intense hypodense artifact, a white background image was used, because the artifact region had the same pixel value as the black background image.

Fig. 2. Image processing for quantitative analysis. A. Image of the tooth in the dental arch. B. Image of the tooth with a black background. C. Image of the tooth after removal of intracanal material (pin and cement). D. Image of the tooth with threshold tool.

ImageJ’s threshold tool was used to identify the hypodense areas of image artifacts within a limited threshold based on the evaluated image pixel values (minimum of 2 and maximum of 159 pixels). The same steps were applied for hyperdense artifact quantification, except that the threshold for hyperdense artifacts were different (minimum of 158 and maximum of 255 pixels).18 Then, the "Analyze > Measure" command was selected, obtaining the standard means and deviations of 1) pixel intensity, which corresponded to the intensity of the grayscales; 2) pixel area, which corresponded to the number of pixels present in the analyzed region (Fig. 3).

Fig. 3. Quantitative analysis using the thresholding tool. A. Image of the tooth after ROI selection. B. Image of the tooth with the threshold tool applied. C. Image of the results after classifying the artifacts into pixels using the threshold tool.

Data analyses

Data were analyzed using IBM SPSS Statistics software version 20.0 (IBM Corp, Armonk, NY, USA). The type of material used for intracanal filling was used as an independent variable. The number of pixels and the area of the image artifacts were considered as response variables. Two-way analysis of variance (ANOVA) was performed to detect statistically significant factors, or interactions between them, that interfered with the response variables. Multiple comparisons of means were made using Tukey HSD post-hoc tests. For the qualitative analysis of the artifacts, Fisher’s exact test was used to compare the proportions of artifacts between groups (intergroup) and the McNemar test to compare the proportions within groups (intragroup) when evaluating non-HD and HD images. In all analyses, the level of significance was set at 5% (P<0.05).

Results

In the qualitative analysis, statistically significant differences were identified between the filling materials (P<0.05). Considering both HD and non-HD images, the percentage of artifacts classified as severe was higher in groups G3 and G6. Additionally, the percentage of artifacts classified as severe significantly increased following the evaluation of HD images, particularly in group G6 (Table 1).

Table 1. Qualitative evaluations regarding the presence of image artifacts according to the type of material and the parameters tested [number (%)].

HD: high definition, ^*: P<0.05 compared between HD and non-HD modes

In the quantitative analysis, no significant differences were found among the parameters studied after accounting for all variables (P>0.05). The mean pixel values and artifact areas did not differ significantly between the groups, nor between the evaluations of images generated with and without HD (Table 2).

Table 2. Quantitative evaluations for the presence of hyperdense and hypodense imaging artifacts, according to the type of material and the parameters tested [mean (standard deviation)].

HD: high definition, Pixel: intensity of grayscale, Area: number of pixels present in the analyzed region. Different letters denote significantly different results (P<0.05). Capital letters indicate horizontal comparisons between modes (HD vs non-HD), while lowercase letters indicate vertical comparisons among groups (G1–G6).

Discussion

It was qualitatively verified that the generation of artifacts exhibited greater severity in groups composed of fiberglass pins reinforced with metallic filaments, regardless of the cement used. The Reforpost^® utilized in this research consists of fiberglass (80%), epoxy resin (19%), and a stainless-steel filament (1%) inside.8,19 This filament is likely responsible for the highest degree of severity observed in the artifacts, as high-density materials, such as the steel present in the pins, tend to produce. Freitas et al.11 also note that high-density materials may be responsible for greater artifact generation.

Chemically, the cementing materials used in this experiment are composed of more than 50% inorganic material (FGM, Joinville, SC, Brazil; Kuraray, Osaka, Japan). Most resin cements contain the following in their composition: Oxygen, Silicon, Carbon, Aluminum, and Barium. The radiopacity of cements relates to elements with high atomic numbers, such as Ytterbium, Lanthanum, Barium, Zirconium, Strontium, and Bromine.20 In this study, it was observed that the cements used did not differ in artifact formation, possibly due to their compounds' low atomic numbers.

In addition to density, another factor that may have contributed to this result is related to the thin layer of cement used, as the pins were pre-modeled in the channel, consequently reducing the cement layer. Possibly, this amount was insufficient to influence artifact generation. Under the conditions applied in this research, it is suggested that dentists can choose any of these cements and may consider additional clinical criteria for material selection, such as longevity and strength.21,22 No previous studies have investigated the influence of fiberglass pins cements on the generation of artifacts. Thus, to reinforce the findings of this study, it is suggested that further research explore the evaluation of the artifacts generated by the cements in other tomographic devices, as well as in other image acquisition procedures parameters.

Regarding exposure parameters, previous studies have shown that this variation does not reduce the production of artifacts.15,18,23 In addition, when assessing how exposure parameters affect the detection of fractures in teeth with various intracanal materials, a study concluded that these parameters also did not significantly impact the reduction of artifacts.23,24 However, the present study found that all groups subjected to the non-HD exposure parameters exhibited a lower degree of artifact severity than those exposed with HD.

In HD mode, 2.5 times more images are created, that is, it produces 500 individual images instead of 200 (non-HD mode) in a single scan, from which the 3D radiography is computed. In HD mode, more images are produced, which enhances quality; however, they involve a higher radiation dose and potentially more artifacts.25 Thus, for fiberglass pins tooth scanning, with or without a metal filament, while using the ORTHOPHOS XG 3D Ready device, the operator can choose to use the fixed non-HD mode that uses 70kV, 50 mAs, 200 image basis to produce a smaller number of artifacts using lower exposure parameters than the HD mode (70kV, 84mAs, 500 images).

Regarding the quantification of hypodense and hyperdense artifacts, no difference in the properties of the materials used was observed. However, a study evaluating the amount of imaging artifacts generated by different intracanal materials in biradicular teeth inferred that those materials with a higher atomic number, such as the fiberglass pin with a steel filament, produced more artifacts.8 These divergences can be explained by methodological differences, such as the number of tooth roots, the computed tomography (CT) device used, and the exposure parameters.

Currently, research is investigating the development of tools that significantly reduce the generation of artifacts; for example, metal artifact reduction (MAR).4,7,25 Dalili et al.26 evaluated the efficacy of CBCT in diagnosing root fractures in the presence of an intracanal pin, both with and without the application of the MAR, concluding that the use of this tool increased the sensitivity for the diagnosis.

Based on this study's results, higher radiation doses do not reduce artifact production; therefore, lower exposure parameters are recommended to minimize patient radiation exposure. To enhance CBCT scans, manufacturers are expected to improve artifact-reduction algorithms and develop new studies to establish image-acquisition protocols that mitigate artifact generation.

This study has limitations regarding its experimental model, which cannot fully reproduce the clinical conditions. However, it would not be possible to perform the evaluations proposed in this study in a clinical trial due to the need to repeat image acquisitions, which would expose the patient to varying radiation doses. The use of a gold-standard examination for the maxillofacial region, as well as the variety of intracanal materials and exposure protocols, are positive characteristics of this experiment. Further studies are suggested to evaluate these materials in multirooted teeth, using other CT scanners to explore new acquisition protocols, since the device studied did not allow the selection of other exposure parameters.

From the results obtained in this research, it was possible to conclude that the cements associated with the fiberglass pins did not influence the production of the artifacts. Furthermore, fiberglass pins with metallic filaments exhibited greater severity in artifact generation. Lower parameter protocols resulted in less artifact production in the assessed CBCT device scans.