Division of Oral Radiology, Department of Oral Diagnosis, Piracicaba Dental School, University of Campinas (UNICAMP), Piracicaba, Brazil.
Division of Oral Radiology, Department of Stomatology, Public Oral Health, Forensic Dentistry, School of Dentistry of Ribeirao Preto, University of Sao Paulo (USP), Ribeirao Preto, Brazil.
Int Endod J. 2020 Sep;53(9):1299-1306. doi: 10.1111/iej.13348. Epub 2020 Jul 9.
To assess the influence of artefacts generated by metal posts on the detection of simulated internal root resorption (IRR) in adjacent teeth using cone-beam computed tomography (CBCT) and to verify the impact of metal artefact reduction (MAR) on these cases.
CBCT images of 14 premolar teeth were acquired before and after IRR simulation using chemical and mechanical procedures, in an OP300 Maxio unit, with and without MAR. Each tooth was placed in the socket of a human mandible and scanned under three different conditions: (i) without adjacent teeth - control group; (ii) distal adjacent tooth restored with metal post; and (iii) with both adjacent teeth restored with metal post. Five oral radiologists scored the IRR detection using a 5-point scale. Diagnostic values were obtained for the tested groups and compared using two-way analysis of variance (α = 0.05).
The presence of a single adjacent tooth restored with metal post did not significantly influence the diagnostic values for IRR detection (P > 0.05). The presence of both adjacent teeth with metal posts, without MAR application, was associated with a significantly lower area under the ROC curve (Az) compared to the control (P = 0.0182). In this case, the application of MAR increased Az, leading to nonsignificant differences from the control group and the group with one adjacent restored tooth (P > 0.05). Sensitivity decreased significantly when two adjacent restored teeth were present, regardless of MAR application (P = 0.0379). Specificity was not affected by the conditions tested (P > 0.05).
CBCT detection of IRR was impaired by artefacts only when both adjacent teeth restored with metal posts were present. In such cases, activation of MAR improved the performance on this diagnostic task.
评估金属桩产生的伪影对使用锥形束 CT（CBCT）检测相邻牙齿模拟内吸收（IRR）的影响，并验证金属伪影减少（MAR）对这些病例的影响。
使用化学和机械程序在 OP300 Maxio 单元中对 14 颗前磨牙进行 IRR 模拟前后的 CBCT 图像采集，有和没有 MAR。将每颗牙齿放置在人类下颌骨的牙槽中，并在三种不同条件下进行扫描：（i）无相邻牙齿 - 对照组；（ii）用金属桩修复的远中相邻牙齿；（iii）用金属桩修复的两个相邻牙齿。五位口腔放射科医生使用 5 分制对 IRR 检测进行评分。对测试组获得诊断值，并使用双向方差分析（α=0.05）进行比较。
单个用金属桩修复的相邻牙齿的存在并未显著影响 IRR 检测的诊断值（P>0.05）。没有应用 MAR 时，两个相邻牙齿都用金属桩修复，与对照组相比，ROC 曲线下面积（Az）显著降低（P=0.0182）。在这种情况下，应用 MAR 增加了 Az，导致与对照组和一个相邻修复牙齿的组无显著差异（P>0.05）。无论是否应用 MAR，当存在两个相邻修复的牙齿时，敏感性显著降低（P=0.0379）。特异性不受测试条件的影响（P>0.05）。
只有当两个相邻牙齿都用金属桩修复时，CBCT 才会受到伪影的影响而无法检测到 IRR。在这种情况下，激活 MAR 可以改善这项诊断任务的性能。
