In vitro assessment of artifacts induced by titanium dental implants in cone beam computed tomography

The objective of this study was to evaluate the geometric pattern and the intensity of artifacts around titanium implants in cone beam computed tomography (CBCT) using an in vitro model.

Ten test models, each containing one 4.1-mm-diameter titanium implant, were cast from a human mandible using silicone impression material and dental stone. Each model contained an implant in one of the following single-tooth gaps: 37, 36, 34, 33, 31, 41, 43, 44, 46, and 47. For control purposes, three models without implants were produced. Each model was scanned five times using a CBCT scanner. Gray values (GV) were recorded at eight circumferential positions around the implants at 0.5 mm, 1 mm, and 2 mm from the implant surface (GVTest ). GV were measured in the corresponding volumes of interest (VOI) in the models without implants (GVControl ). Differences of gray values (ΔGV) between GVTest and GVControl were calculated as percentages. To detect differences between GVTest and GVControl , the 95% confidence interval (CI) was computed for the values of ΔGV. Repeated measures ANOVA was used for the comparison of ΔGV at 0.5 mm, 1 mm, and 2 mm from the implant surface.

Artifacts reflected by altered GV were always present in the proximity of titanium implants, regardless of the implant position. When comparing GVTest and GVControl , increased GV were found at the buccal and lingual aspects of the implant sites, whereas regions with reduced GV were located along the long axis of the mandibular body of the test models. A significant decrease in artifact intensity was found with increasing distance from the buccal implant surface (ΔGV0.5 mm : 45 ± 10% [SD], ΔGV1 mm : 28 ± 14% [SD], ΔGV2 mm : 14 ± 7% [SD]) (P < 0.001).

Artifacts around titanium implants in CBCT images were distributed according to a geometrical pattern.