Accuracy of cone-beam computed tomography for the evaluation of mandible invasion by oral squamous cell carcinoma
Abstract Purpose: Surgery for oral cancer with mandibular invasion requires an accurate preoperative assessment. Although cone-beam computed tomography (CBCT), which yields high spatial resolution, is used widely for mandibular examinations, evidence supporting its accuracy for the evaluation of malignant mandibular invasion is insufficient. Therefore, this study aimed to explore the accuracy of CBCT for the preoperative assessment of mandibular invasion. Methods: Thirty mandibular specimens acquired via mandibulectomy for oral squamous cell carcinoma (OSCC) treatment were collected. The excised samples were marked and subjected to CBCT imaging. Hematoxylin–eosin staining was used for histopathological assessment. Mandibular invasion was estimated based on CBCT and pathologic findings. Invasive borders were delineated via histopathological examination and CBCT and merged to compare the extent of invasion. Results: Although CBCT predicted the excised mandible invasion with 100% accuracy, it predicted nerve invasion with only 69.2% accuracy. The bone samples exhibited significant shrinkage (8.9%) due to tissue processing. This bone mass shrinkage led to a tendency of CBCT to underestimate the extent of invasion, with a mean difference of 2.97 mm. Conclusion: CBCT can reliably assess mandibular invasion, but is much less accurate for the estimation of nerve invasion. Given the potential for underestimation of the extent of mandibular invasion, CBCT would be better used to determine the extent of osteotomy.