Effect of metal artifact reduction level on the assessment of dental implant positioning by cone-beam computed tomography.

Abstract

This study evaluated the effect of metal artifact reduction (MAR) level and tube current on the assessment of dental implant positioning relative to the mandibular canal (MC) through cone-beam computed tomography (CBCT). Titanium dental implants were placed in dried mandibles at 0.5 mm superior to the MC (Group 1/n = 8), 0.5 mm inside the MC with perforation of the cortex (Group 2/n = 10). CBCT scans were obtained with different levels of MAR (off, medium and high) and two tube currents (4 mA and 8 mA). Four examiners analyzed the images and scored the contact between the implant and the MC using a 5-point scale. Sensitivity, specificity, area under receiver operating characteristic curve (ROC) and frequency of scores were calculated. Data were compared with ANOVA two-way and Tukey's test and scores with Chi-Square test. Specificity and area under ROC curve decreased significantly when MAR level was high compared with MAR-medium and MAR-off. The frequency of score 3 (inconclusive) was the highest, and scores 1 and 5 (definitely no contact and definitely contact, respectively) were the lowest with MAR-high, regardless of the tube current. When MAR was off, there were higher frequencies of scores 1 and 5. The level of MAR influences the assessment of the relationship between the dental implant and the MC. MAR-high led to lower diagnostic accuracy compared with MAR-medium and off. This paper shows that high level of MAR can interfere in the diagnostic of dental implant positioning relative to the mandibular canal, decreasing its accuracy.