Objectives: to evaluate the influence of the metal artifact reduction algorithm (MAR) on detecting root perforation adjacent to titanium implant using CBCT.
Methodology: This observational analytical ex vivo study was conducted on 56 human teeth. Two titanium implants were inserted in a dry human mandible at the site of the lower right and left first premolars. Then, decoronated single-rooted endodontically cleaned and shaped teeth were placed in the sockets adjacent to the implants. Of the 56 teeth, 28 were without perforation (control group), and 28 had simulated root perforations. CBCT scans were obtained without and with a metal artifact reduction filter and assessed by two dental radiologists and one general practitioner. Results were statistically analyzed in terms of accuracy, specificity, and sensitivity in addition to Kendall's coefficient of concordance and receiver operator characteristic (ROC) curve analysis.
Results: There was a significant difference between the results of acquisitions without and with the metal artifact reduction algorithm (MAR). Acquisitions without MAR had higher values; accuracy (91.1%) and sensitivity (92.9%). In contrast, acquisitions with MAR had lower values; accuracy (71.4%) and sensitivity (50%).
Conclusion: The metal artifact reduction algorithm hinders diagnosing lateral root perforation adjacent to titanium implants.